engram-core 0.21.1

AI Memory Infrastructure - Persistent memory for AI agents with semantic search
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
# MCP Tools Reference

<!-- GENERATED: do not edit manually. Run `./scripts/generate-mcp-reference.sh`. -->

This reference documents the MCP surface that turns Engram into a shared source of truth for team memory.

It is generated from `src/mcp/tools/registry.rs`.

Total tools: **278**

## Summary

| Tool | Tier | Annotations | Required Inputs |
|------|------|-------------|-----------------|
| `memory_create` | essential | mutating (no MCP hints) | `content` |
| `context_seed` | essential | mutating (no MCP hints) | `facts` |
| `memory_seed` | advanced | mutating (no MCP hints) | `facts` |
| `memory_get` | essential | readOnlyHint | `id` |
| `memory_update` | essential | mutating (no MCP hints) | `id` |
| `memory_delete` | essential | destructiveHint | `id` |
| `memory_list` | essential | readOnlyHint | none |
| `memory_create_todo` | standard | mutating (no MCP hints) | `content` |
| `memory_create_issue` | standard | mutating (no MCP hints) | `title` |
| `memory_versions` | advanced | readOnlyHint | `id` |
| `memory_set_expiration` | standard | mutating (no MCP hints) | `id`, `ttl_seconds` |
| `memory_cleanup_expired` | standard | destructiveHint | none |
| `memory_stats` | essential | readOnlyHint | none |
| `memory_export_graph` | standard | readOnlyHint | none |
| `memory_scan_project` | advanced | mutating (no MCP hints) | none |
| `memory_get_project_context` | advanced | readOnlyHint | none |
| `memory_list_instruction_files` | advanced | readOnlyHint | none |
| `memory_ingest_document` | advanced | mutating (no MCP hints) | `path` |
| `dream_run_now` | advanced | idempotentHint | none |
| `dream_create` | advanced | mutating (no MCP hints) | none |
| `dream_get` | advanced | readOnlyHint | `id` |
| `dream_list` | advanced | readOnlyHint | none |
| `dream_cancel` | advanced | idempotentHint | `id` |
| `dream_archive` | advanced | idempotentHint | `id` |
| `dream_candidates_list` | advanced | readOnlyHint | none |
| `dream_candidate_get` | advanced | readOnlyHint | `id` |
| `dream_candidate_review` | advanced | mutating (no MCP hints) | `id`, `review_state` |
| `dream_candidate_apply` | advanced | mutating (no MCP hints) | `id` |
| `dream_eval_run` | advanced | readOnlyHint | none |
| `workspace_list` | essential | readOnlyHint | none |
| `workspace_stats` | standard | readOnlyHint | `workspace` |
| `workspace_move` | standard | mutating (no MCP hints) | `id`, `workspace` |
| `workspace_delete` | advanced | destructiveHint | `workspace` |
| `memory_create_daily` | standard | mutating (no MCP hints) | `content` |
| `memory_score` | standard | mutating (no MCP hints) | `id` |
| `memory_promote` | standard | mutating (no MCP hints) | `id` |
| `memory_decay` | standard | mutating (no MCP hints) | none |
| `memory_explain` | standard | readOnlyHint | `id` |
| `memory_reconcile_conflict` | standard | mutating (no MCP hints) | `id`, `reason` |
| `memory_promote_to_permanent` | standard | mutating (no MCP hints) | `id` |
| `embedding_cache_stats` | advanced | readOnlyHint | none |
| `embedding_cache_clear` | advanced | destructiveHint | none |
| `session_index` | essential | mutating (no MCP hints) | `messages`, `session_id` |
| `session_index_delta` | standard | mutating (no MCP hints) | `messages`, `session_id` |
| `session_get` | standard | readOnlyHint | `session_id` |
| `session_list` | essential | readOnlyHint | none |
| `session_delete` | standard | destructiveHint | `session_id` |
| `identity_create` | essential | mutating (no MCP hints) | `canonical_id`, `display_name` |
| `identity_get` | standard | readOnlyHint | `canonical_id` |
| `identity_update` | standard | mutating (no MCP hints) | `canonical_id` |
| `identity_delete` | standard | destructiveHint | `canonical_id` |
| `identity_add_alias` | standard | mutating (no MCP hints) | `alias`, `canonical_id` |
| `identity_remove_alias` | advanced | mutating (no MCP hints) | `alias` |
| `identity_resolve` | essential | readOnlyHint | `alias` |
| `identity_list` | standard | readOnlyHint | none |
| `identity_search` | standard | readOnlyHint | `query` |
| `identity_link` | standard | mutating (no MCP hints) | `canonical_id`, `memory_id` |
| `identity_unlink` | advanced | mutating (no MCP hints) | `canonical_id`, `memory_id` |
| `memory_get_identities` | advanced | readOnlyHint | `id` |
| `memory_create_batch` | standard | mutating (no MCP hints) | `memories` |
| `memory_delete_batch` | standard | destructiveHint | `ids` |
| `memory_ingest_fact` | standard | mutating (no MCP hints) | `fact` |
| `memory_ingest_fact_batch` | standard | mutating (no MCP hints) | `facts` |
| `memory_tags` | standard | readOnlyHint | none |
| `memory_tag_hierarchy` | advanced | readOnlyHint | none |
| `memory_validate_tags` | advanced | readOnlyHint | none |
| `memory_rebuild_embeddings` | advanced | idempotentHint | none |
| `memory_rebuild_crossrefs` | advanced | idempotentHint | none |
| `context_budget_check` | advanced | readOnlyHint | `budget`, `memory_ids`, `model` |
| `pending_injections_count` | advanced | readOnlyHint | none |
| `pending_injections_cleanup` | advanced | destructiveHint | none |
| `memory_archive_old` | advanced | destructiveHint | none |
| `memory_from_trace` | advanced | mutating (no MCP hints) | `trace_id` |
| `lifecycle_status` | standard | readOnlyHint | none |
| `lifecycle_run` | standard | idempotentHint | none |
| `memory_set_lifecycle` | advanced | mutating (no MCP hints) | `id`, `state` |
| `lifecycle_config` | advanced | readOnlyHint | none |
| `retention_policy_set` | standard | mutating (no MCP hints) | `workspace` |
| `retention_policy_get` | standard | readOnlyHint | `workspace` |
| `retention_policy_list` | standard | readOnlyHint | none |
| `retention_policy_delete` | advanced | destructiveHint | `workspace` |
| `retention_policy_apply` | advanced | idempotentHint | none |
| `salience_get` | advanced | readOnlyHint | `id` |
| `salience_set_importance` | advanced | mutating (no MCP hints) | `id`, `importance` |
| `salience_boost` | advanced | mutating (no MCP hints) | `id` |
| `salience_demote` | advanced | mutating (no MCP hints) | `id` |
| `salience_decay_run` | advanced | destructiveHint | none |
| `salience_stats` | advanced | readOnlyHint | none |
| `salience_history` | advanced | readOnlyHint | `id` |
| `salience_top` | advanced | readOnlyHint | none |
| `quality_score` | standard | readOnlyHint | `id` |
| `quality_report` | standard | readOnlyHint | none |
| `quality_find_duplicates` | advanced | readOnlyHint | none |
| `quality_get_duplicates` | advanced | readOnlyHint | none |
| `quality_find_conflicts` | advanced | readOnlyHint | `id` |
| `quality_get_conflicts` | advanced | readOnlyHint | none |
| `quality_resolve_conflict` | advanced | destructiveHint | `conflict_id`, `resolution` |
| `quality_source_trust` | advanced | readOnlyHint | `source_type` |
| `quality_improve` | advanced | mutating (no MCP hints) | `id` |
| `memory_export_markdown` | advanced | readOnlyHint | `workspace` |
| `memory_import_markdown` | advanced | mutating (no MCP hints) | `input_dir` |
| `memory_block_create` | standard | mutating (no MCP hints) | `name` |
| `memory_block_get` | standard | readOnlyHint | `name` |
| `memory_block_edit` | standard | mutating (no MCP hints) | `content`, `name` |
| `memory_block_list` | standard | readOnlyHint | none |
| `memory_block_archive` | standard | mutating (no MCP hints) | `name` |
| `memory_block_history` | standard | readOnlyHint | `name` |
| `memory_cache_stats` | standard | readOnlyHint | none |
| `memory_cache_clear` | advanced | mutating (no MCP hints) | none |
| `memory_compress_for_context` | standard | readOnlyHint | `ids` |
| `memory_embedding_migrate` | advanced | mutating (no MCP hints) | none |
| `memory_embedding_providers` | standard | readOnlyHint | none |
| `temporal_add_edge` | advanced | mutating (no MCP hints) | `from_id`, `relation`, `to_id`, `valid_from` |
| `temporal_contradictions` | advanced | readOnlyHint | none |
| `temporal_diff` | advanced | readOnlyHint | `t1`, `t2` |
| `temporal_snapshot` | advanced | readOnlyHint | `timestamp` |
| `temporal_timeline` | advanced | readOnlyHint | `from_id`, `to_id` |
| `memory_enrichment_timeline` | standard | readOnlyHint | `memory_id` |
| `memory_enrichment_audit` | advanced | readOnlyHint | none |
| `memory_search` | essential | readOnlyHint | `query` |
| `memory_smart_retrieve` | essential | readOnlyHint | `query` |
| `memory_digest` | essential | readOnlyHint | `topic` |
| `memory_council` | standard | mutating (no MCP hints) | `prompt` |
| `memory_search_suggest` | standard | readOnlyHint | `query` |
| `memory_find_duplicates` | standard | readOnlyHint | none |
| `memory_find_semantic_duplicates` | standard | readOnlyHint | none |
| `langfuse_connect` | advanced | mutating (no MCP hints) | none |
| `langfuse_sync` | advanced | mutating (no MCP hints) | none |
| `langfuse_sync_status` | advanced | readOnlyHint | `task_id` |
| `langfuse_extract_patterns` | advanced | readOnlyHint | none |
| `search_cache_feedback` | standard | mutating (no MCP hints) | `positive`, `query` |
| `search_cache_stats` | advanced | readOnlyHint | none |
| `search_cache_clear` | advanced | destructiveHint | none |
| `memory_search_by_identity` | standard | mutating (no MCP hints) | `identity` |
| `memory_session_search` | standard | mutating (no MCP hints) | `query` |
| `meilisearch_search` | advanced | readOnlyHint | `query` |
| `meilisearch_reindex` | advanced | idempotentHint | none |
| `meilisearch_status` | advanced | readOnlyHint | none |
| `meilisearch_config` | advanced | readOnlyHint | none |
| `memory_search_compact` | essential | readOnlyHint | `query` |
| `memory_expand` | essential | readOnlyHint | `ids` |
| `memory_detect_updates` | standard | readOnlyHint | `content` |
| `memory_explain_search` | standard | readOnlyHint | `results` |
| `memory_suggest_acquisitions` | standard | readOnlyHint | none |
| `memory_link` | essential | mutating (no MCP hints) | `from_id`, `to_id` |
| `memory_unlink` | standard | mutating (no MCP hints) | `from_id`, `to_id` |
| `memory_related` | essential | readOnlyHint | `id` |
| `memory_extract_entities` | standard | idempotentHint | `id` |
| `memory_get_entities` | standard | readOnlyHint | `id` |
| `memory_search_entities` | standard | readOnlyHint | `query` |
| `memory_entity_stats` | advanced | readOnlyHint | none |
| `memory_traverse` | essential | readOnlyHint | `id` |
| `memory_find_path` | standard | readOnlyHint | `from_id`, `to_id` |
| `memory_graph_path` | advanced | readOnlyHint, idempotentHint | `scope`, `source_id`, `target_id` |
| `memory_temporal_snapshot` | advanced | readOnlyHint, idempotentHint | `scope`, `timestamp` |
| `memory_scope_snapshot` | advanced | readOnlyHint, idempotentHint | `from_timestamp`, `scope`, `to_timestamp` |
| `memory_auto_link` | advanced | mutating (no MCP hints) | none |
| `memory_auto_link_stats` | standard | readOnlyHint | none |
| `memory_cluster` | advanced | readOnlyHint | none |
| `memory_coactivation_report` | standard | readOnlyHint | none |
| `memory_fact_graph` | standard | readOnlyHint | `subject` |
| `memory_garden` | advanced | mutating (no MCP hints) | none |
| `memory_garden_preview` | standard | readOnlyHint | none |
| `memory_get_cluster` | standard | readOnlyHint | `memory_id` |
| `memory_knowledge_stats` | standard | readOnlyHint | none |
| `memory_list_auto_links` | standard | readOnlyHint | none |
| `memory_list_clusters` | standard | readOnlyHint | none |
| `memory_list_facts` | standard | readOnlyHint | none |
| `memory_query_triplets` | standard | readOnlyHint | none |
| `memory_reflect` | standard | readOnlyHint | `ids` |
| `memory_resolve_conflict` | standard | mutating (no MCP hints) | `conflict_id` |
| `memory_sentiment_analyze` | standard | readOnlyHint | `id` |
| `memory_sentiment_timeline` | standard | readOnlyHint | none |
| `memory_sync_status` | advanced | readOnlyHint | none |
| `memory_sync_media` | advanced | mutating (no MCP hints) | none |
| `memory_events_poll` | advanced | readOnlyHint | none |
| `memory_events_clear` | advanced | destructiveHint | none |
| `sync_version` | advanced | readOnlyHint | none |
| `sync_delta` | advanced | readOnlyHint | `since_version` |
| `sync_state` | advanced | readOnlyHint | `agent_id` |
| `sync_cleanup` | advanced | destructiveHint | none |
| `memory_share` | advanced | mutating (no MCP hints) | `from_agent`, `memory_id`, `to_agent` |
| `memory_shared_poll` | advanced | readOnlyHint | `agent_id` |
| `memory_share_ack` | advanced | mutating (no MCP hints) | `agent_id`, `share_id` |
| `memory_grant_access` | advanced | mutating (no MCP hints) | `agent_id`, `scope_path` |
| `memory_revoke_access` | advanced | destructiveHint | `agent_id`, `scope_path` |
| `memory_list_grants` | advanced | readOnlyHint | `agent_id` |
| `memory_check_access` | advanced | readOnlyHint | `agent_id`, `scope_path` |
| `memory_search_by_image` | advanced | readOnlyHint | `image_path` |
| `memory_upload_image` | advanced | mutating (no MCP hints) | `file_path`, `memory_id` |
| `memory_migrate_images` | advanced | idempotentHint | none |
| `memory_capture_screenshot` | advanced | readOnlyHint | none |
| `memory_describe_image` | advanced | readOnlyHint | `image_path` |
| `memory_list_media` | standard | readOnlyHint | none |
| `memory_process_video` | advanced | mutating (no MCP hints) | `video_path` |
| `memory_transcribe_audio` | advanced | readOnlyHint | `audio_path` |
| `agent_register` | advanced | mutating (no MCP hints) | `agent_id` |
| `agent_deregister` | advanced | destructiveHint | `agent_id` |
| `agent_heartbeat` | advanced | mutating (no MCP hints) | `agent_id` |
| `agent_list` | advanced | readOnlyHint | none |
| `agent_get` | advanced | readOnlyHint | `agent_id` |
| `agent_capabilities` | advanced | mutating (no MCP hints) | `agent_id`, `capabilities` |
| `snapshot_create` | advanced | mutating (no MCP hints) | `output_path` |
| `snapshot_load` | advanced | mutating (no MCP hints) | `path`, `strategy` |
| `snapshot_inspect` | advanced | readOnlyHint | `path` |
| `attestation_log` | advanced | mutating (no MCP hints) | `content`, `document_name` |
| `attestation_verify` | advanced | readOnlyHint | `content` |
| `attestation_chain_verify` | advanced | readOnlyHint | none |
| `attestation_list` | advanced | readOnlyHint | none |
| `memory_agent_start` | standard | readOnlyHint | none |
| `memory_agent_stop` | standard | readOnlyHint | none |
| `memory_agent_status` | standard | readOnlyHint | none |
| `memory_agent_metrics` | advanced | mutating (no MCP hints) | none |
| `harness_record` | advanced | mutating (no MCP hints) | `kind`, `summary` |
| `harness_status` | advanced | readOnlyHint | none |
| `harness_handoff` | advanced | mutating (no MCP hints) | `current_goal`, `next_steps` |
| `harness_verify` | advanced | mutating (no MCP hints) | `command`, `exit_code`, `output_summary` |
| `memory_suggest_tags` | advanced | readOnlyHint | none |
| `memory_auto_tag` | advanced | mutating (no MCP hints) | `id` |
| `session_context_create` | standard | mutating (no MCP hints) | `name` |
| `session_context_add_memory` | advanced | mutating (no MCP hints) | `memory_id`, `session_id` |
| `session_context_remove_memory` | advanced | mutating (no MCP hints) | `memory_id`, `session_id` |
| `session_context_get` | standard | readOnlyHint | `session_id` |
| `session_context_list` | standard | readOnlyHint | none |
| `session_context_search` | standard | readOnlyHint | `query`, `session_id` |
| `session_context_update_summary` | advanced | mutating (no MCP hints) | `session_id`, `summary` |
| `session_context_end` | advanced | mutating (no MCP hints) | `session_id` |
| `session_context_export` | advanced | readOnlyHint | `session_id` |
| `memory_get_public` | advanced | readOnlyHint | `id` |
| `memory_get_injection_prompt` | essential | readOnlyHint | `query` |
| `memory_observe_tool_use` | standard | mutating (no MCP hints) | `tool_input`, `tool_name`, `tool_output` |
| `memory_archive_tool_output` | standard | mutating (no MCP hints) | `raw_output`, `tool_name` |
| `memory_get_archived_output` | standard | readOnlyHint | `archive_id` |
| `memory_get_working_memory` | standard | readOnlyHint | `session_id` |
| `session_land` | essential | mutating (no MCP hints) | `session_id` |
| `memory_build_context` | standard | readOnlyHint | `query` |
| `context_record` | standard | mutating (no MCP hints) | `event_type`, `session_id`, `source` |
| `context_record_artifact` | standard | mutating (no MCP hints) | `kind` |
| `context_get_artifact` | standard | readOnlyHint | `artifact_id`, `reason` |
| `context_search` | standard | readOnlyHint | `query` |
| `context_build_bundle` | standard | readOnlyHint | none |
| `recent_activity` | essential | readOnlyHint | none |
| `discover_tools` | essential | readOnlyHint | none |
| `memory_prepare_context` | advanced | readOnlyHint | `query` |
| `memory_extract_facts` | standard | mutating (no MCP hints) | `memory_id` |
| `scope_get` | standard | readOnlyHint | `memory_id` |
| `scope_list` | standard | readOnlyHint | none |
| `scope_search` | standard | readOnlyHint | `query`, `scope_path` |
| `scope_set` | standard | mutating (no MCP hints) | `memory_id`, `scope_path` |
| `scope_tree` | standard | readOnlyHint | none |
| `memory_soft_trim` | advanced | readOnlyHint | `id` |
| `memory_list_compact` | standard | readOnlyHint | none |
| `memory_content_stats` | advanced | readOnlyHint | `id` |
| `memory_export` | advanced | readOnlyHint | none |
| `memory_import` | advanced | mutating (no MCP hints) | `data` |
| `memory_create_section` | standard | mutating (no MCP hints) | `title` |
| `memory_checkpoint` | standard | mutating (no MCP hints) | `session_id`, `summary` |
| `memory_create_episodic` | standard | mutating (no MCP hints) | `content`, `event_time` |
| `memory_create_procedural` | standard | mutating (no MCP hints) | `content`, `trigger_pattern` |
| `memory_get_timeline` | standard | readOnlyHint | none |
| `memory_get_procedures` | standard | readOnlyHint | none |
| `memory_record_procedure_outcome` | standard | mutating (no MCP hints) | `id`, `success` |
| `memory_boost` | standard | mutating (no MCP hints) | `id` |
| `memory_explain_utility` | standard | readOnlyHint | `memory_id` |
| `memory_summarize` | advanced | mutating (no MCP hints) | `memory_ids` |
| `memory_get_full` | advanced | readOnlyHint | `id` |
| `memory_auto_consolidate` | advanced | mutating (no MCP hints) | `action` |
| `memory_consolidate_batch` | advanced | destructiveHint | none |
| `memory_consolidation_history` | advanced | readOnlyHint | none |
| `memory_compress` | advanced | readOnlyHint | `id` |
| `memory_consolidate` | advanced | mutating (no MCP hints) | none |
| `memory_decompress` | standard | readOnlyHint | `id` |
| `memory_detect_conflicts` | standard | mutating (no MCP hints) | none |
| `memory_feedback` | standard | mutating (no MCP hints) | `memory_id`, `query`, `signal` |
| `memory_feedback_stats` | standard | readOnlyHint | none |
| `memory_synthesis` | standard | readOnlyHint | `content_a`, `content_b` |
| `memory_utility_score` | standard | readOnlyHint | `id` |
| `memory_replay_at_time` | advanced | readOnlyHint | `memory_id`, `timestamp` |

## Tools

### `memory_create`

Store an explicit durable memory with inspectable provenance. Use for stable preferences, decisions, insights, and project context when the fact is intentional and worth preserving.

- Tier: `essential`
- Annotations: mutating (no MCP hints)
- Required inputs: `content`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | The content to remember |
| `memory_type` | `string` | no | Memory type (preferred field; alias: type) Default: `note`. Allowed: `note`, `todo`, `issue`, `decision`, `preference`, `learning`, `context`, `credential`, `episodic`, `procedural`, `summary`, `checkpoint`, `image`, `audio`, `video`. |
| `type` | `string` | no | Deprecated alias for memory_type Default: `note`. Allowed: `note`, `todo`, `issue`, `decision`, `preference`, `learning`, `context`, `credential`, `episodic`, `procedural`, `summary`, `checkpoint`, `image`, `audio`, `video`. |
| `tags` | `array` | no | Tags for categorization Items: `string`. |
| `metadata` | `object` | no | Additional metadata as key-value pairs |
| `importance` | `number` | no | Importance score (0-1) Minimum: `0`. Maximum: `1`. |
| `workspace` | `string` | no | Workspace to store the memory in (default: 'default') |
| `tier` | `string` | no | Memory tier: permanent (never expires) or daily (auto-expires) Default: `permanent`. Allowed: `permanent`, `daily`. |
| `defer_embedding` | `boolean` | no | Defer embedding to background queue Default: `false`. |
| `ttl_seconds` | `integer` | no | Time-to-live in seconds. Memory will auto-expire after this duration. Omit for permanent storage. Setting this implies tier='daily'. |
| `dedup_mode` | `string` | no | How to handle duplicate content: reject (error if exact match), merge (combine tags/metadata with existing), skip (return existing unchanged), allow (create duplicate) Default: `allow`. Allowed: `reject`, `merge`, `skip`, `allow`. |
| `dedup_threshold` | `number` | no | Similarity threshold for semantic deduplication (0.0-1.0). When set with dedup_mode != 'allow', memories with cosine similarity >= threshold are treated as duplicates. Requires embeddings. If not set, only exact content hash matching is used. Minimum: `0`. Maximum: `1`. |
| `event_time` | `string` | no | ISO8601 timestamp for episodic memories (when the event occurred) Format: `date-time`. |
| `event_duration_seconds` | `integer` | no | Duration of the event in seconds (for episodic memories) |
| `trigger_pattern` | `string` | no | Pattern that triggers this procedure (for procedural memories) |
| `summary_of_id` | `integer` | no | ID of the memory this summarizes (for summary memories) |
| `media_url` | `string` | no | URL or local path to the primary media asset (for Image/Audio/Video memory types). Format: local:///path, https://..., or s3://... |

### `context_seed`

Injects initial context (premises, persona assumptions, or structured facts) about an entity to avoid cold start. Seeded memories are tagged as origin:seed and status:unverified, and should be treated as revisable assumptions.

- Tier: `essential`
- Annotations: mutating (no MCP hints)
- Required inputs: `facts`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `entity_context` | `string` | no | Name or ID of the entity (e.g., 'Client: Roberto', 'Account: ACME', 'Project: Alpha') Max length: `200`. |
| `workspace` | `string` | no | Workspace to store the memories in (default: 'default') |
| `base_tags` | `array` | no | Tags applied to all facts (e.g., ['vip', 'prospect']) Items: `string`. |
| `ttl_seconds` | `integer` | no | Override TTL for all facts in seconds (0 = disable TTL). If omitted, TTL is derived from confidence. |
| `disable_ttl` | `boolean` | no | Disable TTL and keep seeded memories permanent regardless of confidence. Default: `false`. |
| `facts` | `array` | yes | Items: `object`. Min items: `1`. |

### `memory_seed`

Deprecated alias for context_seed. Use context_seed instead.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `facts`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `entity_context` | `string` | no | Name or ID of the entity (e.g., 'Client: Roberto', 'Account: ACME', 'Project: Alpha') Max length: `200`. |
| `workspace` | `string` | no | Workspace to store the memories in (default: 'default') |
| `base_tags` | `array` | no | Tags applied to all facts (e.g., ['vip', 'prospect']) Items: `string`. |
| `ttl_seconds` | `integer` | no | Override TTL for all facts in seconds (0 = disable TTL). If omitted, TTL is derived from confidence. |
| `disable_ttl` | `boolean` | no | Disable TTL and keep seeded memories permanent regardless of confidence. Default: `false`. |
| `facts` | `array` | yes | Items: `object`. Min items: `1`. |

### `memory_get`

Retrieve a memory by its ID

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |
| `strip_private` | `boolean` | no | When true, removes all <private>...</private> tagged sections from the content before returning (default: false) |

### `memory_update`

Update an existing memory

- Tier: `essential`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |
| `content` | `string` | no | New content |
| `memory_type` | `string` | no | Memory type (preferred field; alias: type) Allowed: `note`, `todo`, `issue`, `decision`, `preference`, `learning`, `context`, `credential`, `episodic`, `procedural`, `summary`, `checkpoint`, `image`, `audio`, `video`. |
| `type` | `string` | no | Deprecated alias for memory_type Allowed: `note`, `todo`, `issue`, `decision`, `preference`, `learning`, `context`, `credential`, `episodic`, `procedural`, `summary`, `checkpoint`, `image`, `audio`, `video`. |
| `tags` | `array` | no | Items: `string`. |
| `metadata` | `object` | no | No description. |
| `importance` | `number` | no | Minimum: `0`. Maximum: `1`. |
| `ttl_seconds` | `integer` | no | Time-to-live in seconds (0 = remove expiration, positive = set new expiration) |
| `event_time` | `string \| null` | no | ISO8601 timestamp for episodic memories (null to clear) Format: `date-time`. |
| `trigger_pattern` | `string \| null` | no | Pattern that triggers this procedure (null to clear) |
| `media_url` | `string \| null` | no | URL or local path to the primary media asset (null to clear) |

### `memory_delete`

Delete a memory (soft delete)

- Tier: `essential`
- Annotations: destructiveHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |
| `cascade_chain` | `boolean` | no | When true, also delete all memories in the supersedes chain (ancestors this memory replaced). Default: `false`. |

### `memory_list`

List memories with filtering and pagination. Supports workspace isolation, tier filtering, and advanced filter syntax with AND/OR and comparison operators.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `limit` | `integer` | no | Default: `20`. |
| `offset` | `integer` | no | Default: `0`. |
| `tags` | `array` | no | Items: `string`. |
| `memory_type` | `string` | no | Filter by memory type (preferred field; alias: type) |
| `type` | `string` | no | Deprecated alias for memory_type |
| `workspace` | `string` | no | Filter by single workspace |
| `workspaces` | `array` | no | Filter by multiple workspaces Items: `string`. |
| `tier` | `string` | no | Filter by memory tier Allowed: `permanent`, `daily`. |
| `sort_by` | `string` | no | Allowed: `created_at`, `updated_at`, `last_accessed_at`, `importance`, `access_count`. |
| `sort_order` | `string` | no | Default: `desc`. Allowed: `asc`, `desc`. |
| `filter` | `object` | no | Advanced filter with AND/OR logic and comparison operators. Supports workspace, tier, and metadata fields. Example: {"AND": [{"metadata.project": {"eq": "engram"}}, {"importance": {"gte": 0.5}}]}. Supported operators: eq, neq, gt, gte, lt, lte, contains, not_contains, exists. Fields: content, memory_type, importance, tags, workspace, tier, created_at, updated_at, metadata.* |
| `metadata_filter` | `object` | no | Legacy simple key-value filter (deprecated, use 'filter' instead) |

### `memory_create_todo`

Create a TODO memory with priority

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `content`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | No description. |
| `priority` | `string` | no | Default: `medium`. Allowed: `low`, `medium`, `high`, `critical`. |
| `due_date` | `string` | no | Format: `date`. |
| `tags` | `array` | no | Items: `string`. |

### `memory_create_issue`

Create an ISSUE memory for tracking problems

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `title`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `title` | `string` | yes | No description. |
| `description` | `string` | no | No description. |
| `severity` | `string` | no | Default: `medium`. Allowed: `low`, `medium`, `high`, `critical`. |
| `tags` | `array` | no | Items: `string`. |

### `memory_versions`

Get version history for a memory

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | No description. |

### `memory_set_expiration`

Set or update the expiration time for a memory

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`, `ttl_seconds`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |
| `ttl_seconds` | `integer` | yes | Time-to-live in seconds from now. Use 0 to remove expiration (make permanent). |

### `memory_cleanup_expired`

Delete all expired memories. Typically called by a background job, but can be invoked manually.

- Tier: `standard`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_stats`

Get storage statistics

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_export_graph`

Export knowledge graph visualization

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `format` | `string` | no | Default: `html`. Allowed: `html`, `json`. |
| `max_nodes` | `integer` | no | Default: `500`. |
| `focus_id` | `integer` | no | Center graph on this memory |

### `memory_scan_project`

Scan current directory for AI instruction files (CLAUDE.md, AGENTS.md, .cursorrules, etc.) and ingest them as memories. Creates parent memory for each file and child memories for sections.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `path` | `string` | no | Directory to scan (defaults to current working directory) |
| `scan_parents` | `boolean` | no | Also scan parent directories (security: disabled by default) Default: `false`. |
| `extract_sections` | `boolean` | no | Create separate memories for each section Default: `true`. |

### `memory_get_project_context`

Get all project context memories for the current working directory. Returns instruction files and their sections.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `path` | `string` | no | Project path (defaults to current working directory) |
| `include_sections` | `boolean` | no | Include section memories Default: `true`. |
| `file_types` | `array` | no | Filter by file type (claude-md, cursorrules, etc.) Items: `string`. |

### `memory_list_instruction_files`

List AI instruction files (CLAUDE.md, AGENTS.md, .cursorrules, etc.) in a directory without ingesting them. Returns file paths, types, and sizes for discovery purposes.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `path` | `string` | no | Directory to scan (defaults to current working directory) |
| `scan_parents` | `boolean` | no | Also scan parent directories for instruction files Default: `false`. |

### `memory_ingest_document`

Ingest a document (PDF or Markdown) into memory. Extracts text, splits into chunks with overlap, and creates memories with deduplication.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `path` | `string` | yes | Local file path to the document |
| `format` | `string` | no | Document format (auto-detect from extension if not specified) Default: `auto`. Allowed: `auto`, `md`, `pdf`. |
| `chunk_size` | `integer` | no | Maximum characters per chunk Default: `1200`. |
| `overlap` | `integer` | no | Overlap between chunks in characters Default: `200`. |
| `max_file_size` | `integer` | no | Maximum file size in bytes (default 10MB) Default: `10485760`. |
| `tags` | `array` | no | Additional tags to add to all chunks Items: `string`. |

### `dream_run_now`

Manually trigger the Dream Phase (background consolidation) across all workspaces. This process compresses old memories and identifies patterns while the agent is 'sleeping'.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `dream_create`

Create a reviewable dream snapshot job and optionally run deterministic candidate generation. Generated candidates are proposals, not canonical memories.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Default: `default`. |
| `job_id` | `string` | no | Optional stable job id. Omit to generate one. |
| `instructions` | `string` | no | No description. |
| `run` | `boolean` | no | When true, run deterministic generation immediately. Default: `true`. |
| `max_memories` | `integer` | no | Default: `50`. Minimum: `1`. |
| `max_candidates` | `integer` | no | Default: `25`. Minimum: `1`. |
| `summary_min_memories` | `integer` | no | Default: `2`. Minimum: `1`. |

### `dream_get`

Inspect one dream snapshot job.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `string` | yes | Dream job id |

### `dream_list`

List dream snapshot jobs by workspace and status.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | No description. |
| `status` | `string` | no | Allowed: `pending`, `running`, `completed`, `failed`, `canceled`, `archived`. |
| `limit` | `integer` | no | Default: `100`. Minimum: `1`. Maximum: `1000`. |

### `dream_cancel`

Cancel a pending or running dream snapshot job idempotently.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `string` | yes | Dream job id |

### `dream_archive`

Archive a terminal dream snapshot job.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `string` | yes | Dream job id |

### `dream_candidates_list`

List review candidates emitted by dream snapshot jobs. Results are proposals and are not canonical memory facts.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | No description. |
| `job_id` | `string` | no | No description. |
| `review_state` | `string` | no | Allowed: `pending`, `accepted`, `edited`, `rejected`, `applied`, `archived`. |
| `limit` | `integer` | no | Default: `100`. Minimum: `1`. Maximum: `1000`. |

### `dream_candidate_get`

Inspect one dream candidate and its evidence sources.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `string` | yes | Dream candidate id |

### `dream_candidate_review`

Review a dream candidate by accepting, editing, rejecting, or archiving it. This does not mutate canonical memory.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`, `review_state`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `string` | yes | Dream candidate id |
| `review_state` | `string` | yes | Allowed: `accepted`, `edited`, `rejected`, `archived`. |
| `edited_content` | `string` | no | Reviewed replacement content when review_state is edited. |
| `metadata_patch` | `object` | no | Optional review metadata merged into candidate metadata. |

### `dream_candidate_apply`

Apply an accepted or edited dream candidate to canonical memory. Requires confirm=true unless dry_run=true; repeated apply is idempotent.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `string` | yes | Dream candidate id |
| `confirm` | `boolean` | no | Must be true for canonical mutation. Default: `false`. |
| `dry_run` | `boolean` | no | Preview planned canonical mutation without applying. Default: `false`. |

### `dream_eval_run`

Run deterministic local dream snapshot evaluation fixtures and return parseable CI-safe metrics. Does not require network, credentials, or model access.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `fixtures` | `array` | no | Optional subset of fixed fixture names. Omit to run all fixtures. Items: `string`. |
| `include_details` | `boolean` | no | Include per-fixture candidate details. Default: `true`. |

### `workspace_list`

List all workspaces with their statistics (memory count, tier breakdown, etc.)

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `workspace_stats`

Get detailed statistics for a specific workspace

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `workspace`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | yes | Workspace name |

### `workspace_move`

Move a memory to a different workspace

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`, `workspace`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to move |
| `workspace` | `string` | yes | Target workspace name |

### `workspace_delete`

Delete a workspace. Can either move all memories to 'default' workspace or hard delete them.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: `workspace`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | yes | Workspace to delete |
| `move_to_default` | `boolean` | no | If true, moves memories to 'default' workspace. If false, deletes all memories in the workspace. Default: `true`. |

### `memory_create_daily`

Create a daily (ephemeral) memory that auto-expires after the specified TTL. Useful for session context and scratch notes.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `content`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | The content to remember |
| `type` | `string` | no | Default: `note`. Allowed: `note`, `todo`, `issue`, `decision`, `preference`, `learning`, `context`, `credential`. |
| `tags` | `array` | no | Tags for categorization Items: `string`. |
| `metadata` | `object` | no | Additional metadata as key-value pairs |
| `importance` | `number` | no | Importance score (0-1) Minimum: `0`. Maximum: `1`. |
| `ttl_seconds` | `integer` | no | Time-to-live in seconds (default: 24 hours) Default: `86400`. |
| `workspace` | `string` | no | Workspace to store the memory in (default: 'default') |

### `memory_score`

Compute deterministic memory policy scores for a memory. When persist=true, upserts the memory_policy row and emits a best-effort policy audit event.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to score |
| `persist` | `boolean` | no | Persist the computed policy score to memory_policy Default: `false`. |

### `memory_promote`

Reinforce a memory's policy record, optionally promoting a Daily-tier memory to the canonical Permanent tier.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to promote or reinforce |
| `canonical_tier` | `boolean` | no | When true, also call promote_to_permanent for canonical tier promotion Default: `false`. |

### `memory_decay`

Compute or apply conservative memory policy decay for a workspace. Dry-run is the default; apply only updates memory_policy scores and active lifecycle transitions.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to decay Default: `default`. |
| `dry_run` | `boolean` | no | When true, compute candidate changes without mutation Default: `true`. |

### `memory_explain`

Explain a memory's current policy score with feature components, reason text, and policy audit count.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to explain |

### `memory_reconcile_conflict`

Record a conflict reconciliation signal for a memory policy without deleting or mutating memory content.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`, `reason`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID with the conflict signal |
| `reason` | `string` | yes | Audit reason for the conflict reconciliation |

### `memory_promote_to_permanent`

Promote a daily memory to permanent tier. Clears the expiration and makes the memory permanent.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to promote |

### `embedding_cache_stats`

Get statistics about the embedding cache (hits, misses, entries, bytes used, hit rate)

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `embedding_cache_clear`

Clear all entries from the embedding cache

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `session_index`

Index a conversation into searchable memory chunks. Uses dual-limiter chunking (messages + characters) with overlap.

- Tier: `essential`
- Annotations: mutating (no MCP hints)
- Required inputs: `messages`, `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Unique session identifier |
| `messages` | `array` | yes | Array of conversation messages Items: `object`. |
| `title` | `string` | no | Optional session title |
| `workspace` | `string` | no | Workspace to store chunks in (default: 'default') |
| `agent_id` | `string` | no | Optional agent identifier |
| `max_messages` | `integer` | no | Max messages per chunk Default: `10`. |
| `max_chars` | `integer` | no | Max characters per chunk Default: `8000`. |
| `overlap` | `integer` | no | Overlap messages between chunks Default: `2`. |
| `ttl_days` | `integer` | no | TTL for transcript chunks in days Default: `7`. |

### `session_index_delta`

Incrementally index new messages to an existing session. More efficient than full reindex.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `messages`, `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session to update |
| `messages` | `array` | yes | New messages to add Items: `object`. |

### `session_get`

Get information about an indexed session

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID to retrieve |

### `session_list`

List indexed sessions with optional workspace filter

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Filter by workspace |
| `limit` | `integer` | no | Maximum sessions to return Default: `20`. |

### `session_delete`

Delete a session and all its indexed chunks

- Tier: `standard`
- Annotations: destructiveHint
- Required inputs: `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session to delete |

### `identity_create`

Create a new identity with canonical ID, display name, and optional aliases

- Tier: `essential`
- Annotations: mutating (no MCP hints)
- Required inputs: `canonical_id`, `display_name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `canonical_id` | `string` | yes | Unique canonical identifier (e.g., 'user:ronaldo', 'org:acme') |
| `display_name` | `string` | yes | Human-readable display name |
| `entity_type` | `string` | no | Default: `person`. Allowed: `person`, `organization`, `project`, `tool`, `concept`, `other`. |
| `description` | `string` | no | Optional description |
| `aliases` | `array` | no | Initial aliases for this identity Items: `string`. |
| `metadata` | `object` | no | Additional metadata |

### `identity_get`

Get an identity by its canonical ID

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `canonical_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `canonical_id` | `string` | yes | Canonical identifier |

### `identity_update`

Update an identity's display name, description, or type

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `canonical_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `canonical_id` | `string` | yes | Canonical identifier |
| `display_name` | `string` | no | New display name |
| `description` | `string` | no | New description |
| `entity_type` | `string` | no | Allowed: `person`, `organization`, `project`, `tool`, `concept`, `other`. |

### `identity_delete`

Delete an identity and all its aliases

- Tier: `standard`
- Annotations: destructiveHint
- Required inputs: `canonical_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `canonical_id` | `string` | yes | Canonical identifier to delete |

### `identity_add_alias`

Add an alias to an identity. Aliases are normalized (lowercase, trimmed). Conflicts with existing aliases are rejected.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `alias`, `canonical_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `canonical_id` | `string` | yes | Canonical identifier |
| `alias` | `string` | yes | Alias to add |
| `source` | `string` | no | Optional source of the alias (e.g., 'manual', 'extracted') |

### `identity_remove_alias`

Remove an alias from any identity

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `alias`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `alias` | `string` | yes | Alias to remove |

### `identity_resolve`

Resolve an alias to its canonical identity. Returns the identity if found, null otherwise.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `alias`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `alias` | `string` | yes | Alias to resolve |

### `identity_list`

List all identities with optional type filter

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `entity_type` | `string` | no | Allowed: `person`, `organization`, `project`, `tool`, `concept`, `other`. |
| `limit` | `integer` | no | Default: `50`. |

### `identity_search`

Search identities by alias or display name

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query |
| `limit` | `integer` | no | Default: `20`. |

### `identity_link`

Link an identity to a memory (mark that the identity is mentioned in the memory)

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `canonical_id`, `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | Memory ID |
| `canonical_id` | `string` | yes | Identity canonical ID |
| `mention_text` | `string` | no | The text that mentions this identity |

### `identity_unlink`

Remove the link between an identity and a memory

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `canonical_id`, `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | Memory ID |
| `canonical_id` | `string` | yes | Identity canonical ID |

### `memory_get_identities`

Get all identities (persons, organizations, projects, etc.) linked to a memory. Returns identity details including display name, type, aliases, and mention information.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |

### `memory_create_batch`

Create multiple memories in a single operation. More efficient than individual creates for bulk imports.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `memories`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memories` | `array` | yes | Array of memories to create Items: `object`. |

### `memory_delete_batch`

Delete multiple memories in a single operation.

- Tier: `standard`
- Annotations: destructiveHint
- Required inputs: `ids`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `ids` | `array` | yes | Array of memory IDs to delete Items: `integer`. |
| `cascade_chain` | `boolean` | no | When true, also delete all memories in the supersedes chain (ancestors this memory replaced). Default: `false`. |

### `memory_ingest_fact`

Append-only fact ingest for high-frequency sources (sessions, file watchers). Always inserts a new memory with memory_type='fact'. No dedup or upsert.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `fact`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `fact` | `string` | yes | The fact text to store |
| `source` | `string` | no | Origin identifier, e.g. 'session:abc' or 'watcher:/path/to/file' |
| `session_id` | `string` | no | Session ID stored in metadata.session_id |
| `workspace` | `string` | no | Workspace name (default: 'default') |
| `tags` | `array` | no | Optional tags Items: `string`. |
| `importance` | `number` | no | Importance score (default: 0.8) Minimum: `0`. Maximum: `1`. |
| `scope` | `string` | no | Memory scope (default: 'global') |

### `memory_ingest_fact_batch`

Batch append-only fact ingest. Inserts all facts in a single transaction. Returns count and ids.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `facts`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `facts` | `array` | yes | Array of fact objects to insert Items: `object`. |
| `workspace` | `string` | no | Default workspace applied to all facts (default: 'default') |
| `scope` | `string` | no | Memory scope applied to all facts (default: 'global') |

### `memory_tags`

List all tags with usage counts and most recent usage timestamps.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_tag_hierarchy`

Get tags organized in a hierarchical tree structure. Tags with slashes are treated as paths (e.g., 'project/engram/core').

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_validate_tags`

Validate tag consistency across memories. Reports orphaned tags, unused tags, and suggested normalizations.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_rebuild_embeddings`

Rebuild embeddings for all memories that are missing them. Useful after model changes or data recovery.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_rebuild_crossrefs`

Rebuild cross-reference links between memories. Re-analyzes all memories to find and create links.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `context_budget_check`

Check token usage of memories against a budget. Returns token counts and suggestions if over budget.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `budget`, `memory_ids`, `model`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_ids` | `array` | yes | IDs of memories to check Items: `integer`. |
| `model` | `string` | yes | Model name for tokenization (gpt-4, gpt-4o, gpt-4o-mini, claude-3-opus, etc.) |
| `encoding` | `string` | no | Override encoding (cl100k_base, o200k_base). Optional if model is known. |
| `budget` | `integer` | yes | Token budget to check against |

### `pending_injections_count`

Count of non-expired payloads queued in pending_injections for a workspace, waiting to be consumed by the next SessionStart.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Default: `default`. |

### `pending_injections_cleanup`

Drop every pending_injections row whose expires_at has passed. Idempotent. Returns the count removed.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_archive_old`

Archive old, low-importance memories by creating summaries. Moves originals to archived state.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `max_age_days` | `integer` | no | Archive memories older than this many days Default: `90`. |
| `max_importance` | `number` | no | Only archive memories with importance below this Default: `0.5`. |
| `min_access_count` | `integer` | no | Skip memories accessed more than this many times Default: `5`. |
| `workspace` | `string` | no | Limit to specific workspace |
| `dry_run` | `boolean` | no | If true, only report what would be archived Default: `true`. |

### `memory_from_trace`

Create a memory from a specific Langfuse trace ID.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `trace_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `trace_id` | `string` | yes | Langfuse trace ID |
| `memory_type` | `string` | no | Type of memory to create Default: `episodic`. Allowed: `note`, `episodic`, `procedural`, `learning`. |
| `workspace` | `string` | no | Workspace for the memory |
| `tags` | `array` | no | Additional tags Items: `string`. |

### `lifecycle_status`

Get lifecycle statistics (active/stale/archived counts by workspace).

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Filter by workspace (optional) |

### `lifecycle_run`

Manually trigger a lifecycle cycle (mark stale, archive old). Dry run by default.

- Tier: `standard`
- Annotations: idempotentHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `dry_run` | `boolean` | no | Preview changes without applying Default: `true`. |
| `workspace` | `string` | no | Limit to specific workspace |
| `stale_days` | `integer` | no | Mark memories older than this as stale Default: `30`. |
| `archive_days` | `integer` | no | Archive memories older than this Default: `90`. |
| `min_importance` | `number` | no | Only process memories below this importance Default: `0.5`. |

### `memory_set_lifecycle`

Manually set the lifecycle state of a memory.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`, `state`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |
| `state` | `string` | yes | New lifecycle state Allowed: `active`, `stale`, `archived`. |

### `lifecycle_config`

Get or set lifecycle configuration (intervals, thresholds).

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `stale_days` | `integer` | no | Days before marking as stale |
| `archive_days` | `integer` | no | Days before auto-archiving |
| `min_importance` | `number` | no | Importance threshold for lifecycle |
| `min_access_count` | `integer` | no | Access count threshold |

### `retention_policy_set`

Set a retention policy for a workspace. Controls auto-compression, max memory count, and auto-deletion.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `workspace`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | yes | Workspace name |
| `max_age_days` | `integer` | no | Hard age limit — auto-delete after this many days |
| `max_memories` | `integer` | no | Maximum active memories in this workspace |
| `compress_after_days` | `integer` | no | Auto-compress memories older than this |
| `compress_max_importance` | `number` | no | Only compress memories with importance <= this (default 0.3) |
| `compress_min_access` | `integer` | no | Skip compression if access_count >= this (default 3) |
| `auto_delete_after_days` | `integer` | no | Auto-delete archived memories older than this |
| `exclude_types` | `array` | no | Memory types exempt from policy (e.g. ["decision", "checkpoint"]) Items: `string`. |

### `retention_policy_get`

Get the retention policy for a workspace.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `workspace`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | yes | Workspace name |

### `retention_policy_list`

List all retention policies across all workspaces.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `retention_policy_delete`

Delete a retention policy for a workspace.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: `workspace`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | yes | Workspace name |

### `retention_policy_apply`

Apply all retention policies now. Compresses, caps, and deletes per workspace rules.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `dry_run` | `boolean` | no | Preview what would happen without making changes Default: `false`. |

### `salience_get`

Get the salience score for a memory. Returns recency, frequency, importance, and feedback components with the combined score and lifecycle state.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to get salience for |
| `feedback_signal` | `number` | no | Optional feedback signal (-1 to 1) to include in calculation Default: `0`. Minimum: `-1`. Maximum: `1`. |

### `salience_set_importance`

Set the importance score for a memory. This is the static importance component of salience.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`, `importance`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |
| `importance` | `number` | yes | Importance score (0-1) Minimum: `0`. Maximum: `1`. |

### `salience_boost`

Boost a memory's salience score temporarily or permanently. Useful for marking memories as contextually relevant.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to boost |
| `boost_amount` | `number` | no | Amount to boost (0-1) Default: `0.2`. Minimum: `0`. Maximum: `1`. |
| `reason` | `string` | no | Optional reason for boosting |

### `salience_demote`

Demote a memory's salience score. Useful for marking memories as less relevant.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to demote |
| `demote_amount` | `number` | no | Amount to demote (0-1) Default: `0.2`. Minimum: `0`. Maximum: `1`. |
| `reason` | `string` | no | Optional reason for demoting |

### `salience_decay_run`

Run temporal decay on all memories. Updates lifecycle states (Active → Stale → Archived) based on salience scores.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `dry_run` | `boolean` | no | If true, compute changes without persisting updates Default: `false`. |
| `record_history` | `boolean` | no | Record salience history entries while updating Default: `true`. |
| `workspace` | `string` | no | Limit to specific workspace |
| `stale_threshold_days` | `integer` | no | Days of inactivity before marking stale Minimum: `1`. |
| `archive_threshold_days` | `integer` | no | Days of inactivity before suggesting archive Minimum: `1`. |

### `salience_stats`

Get salience statistics across all memories. Returns distribution, percentiles, and state counts.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Limit to specific workspace |

### `salience_history`

Get salience score history for a memory. Shows how salience has changed over time.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |
| `limit` | `integer` | no | Maximum history entries to return Default: `50`. |

### `salience_top`

Get top memories by salience score. Useful for context injection.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `limit` | `integer` | no | Maximum memories to return Default: `20`. |
| `workspace` | `string` | no | Limit to specific workspace |
| `min_score` | `number` | no | Minimum salience score Minimum: `0`. Maximum: `1`. |
| `memory_type` | `string` | no | Filter by memory type |

### `quality_score`

Get the quality score for a memory with detailed breakdown of clarity, completeness, freshness, consistency, and source trust components.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to score |

### `quality_report`

Generate a comprehensive quality report for a workspace. Includes quality distribution, top issues, conflict and duplicate counts.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to analyze (default: 'default') |

### `quality_find_duplicates`

Find near-duplicate memories using text similarity. Returns pairs of similar memories above the threshold.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `threshold` | `number` | no | Similarity threshold (0-1) Default: `0.85`. Minimum: `0`. Maximum: `1`. |
| `limit` | `integer` | no | Maximum memories to compare Default: `100`. |

### `quality_get_duplicates`

Get pending duplicate candidates that need review.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `limit` | `integer` | no | Maximum duplicates to return Default: `50`. |

### `quality_find_conflicts`

Detect conflicts for a memory against existing memories. Finds contradictions, staleness, and semantic overlaps.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to check for conflicts |

### `quality_get_conflicts`

Get unresolved conflicts that need attention.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `limit` | `integer` | no | Maximum conflicts to return Default: `50`. |

### `quality_resolve_conflict`

Resolve a conflict between memories. Options: keep_a, keep_b, merge, keep_both, delete_both, false_positive.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: `conflict_id`, `resolution`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `conflict_id` | `integer` | yes | Conflict ID to resolve |
| `resolution` | `string` | yes | How to resolve the conflict Allowed: `keep_a`, `keep_b`, `merge`, `keep_both`, `delete_both`, `false_positive`. |
| `notes` | `string` | no | Optional notes about the resolution |

### `quality_source_trust`

Get or update trust score for a source type. Higher trust means memories from this source are weighted more in quality calculations.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `source_type`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `source_type` | `string` | yes | Source type (user, seed, extraction, inference, external) |
| `source_identifier` | `string` | no | Optional specific source identifier |
| `trust_score` | `number` | no | New trust score (omit to just get current score) Minimum: `0`. Maximum: `1`. |
| `notes` | `string` | no | Notes about this source |

### `quality_improve`

Get suggestions for improving a memory's quality. Returns actionable recommendations.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to analyze |

### `memory_export_markdown`

Export a workspace as human-readable Markdown files with YAML frontmatter and wiki-style [[links]]. Creates one .md file per memory, organized by type in subdirectories, with an index.md overview.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `workspace`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | yes | Workspace to export |
| `output_dir` | `string` | no | Output directory path (default: ./engram-export/{workspace}/) |
| `include_links` | `boolean` | no | Include [[wiki links]] to related memories in each file Default: `true`. |

### `memory_import_markdown`

Import memories from Markdown files with engram_ frontmatter (RFC 0004). Review mode by default (confirm: false) — returns a staged list without writing. Detects drift via content_hash and version conflicts via engram_version. Ignores non-engram_ frontmatter keys (Obsidian-safe).

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `input_dir`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `input_dir` | `string` | yes | Directory to scan recursively for .md files |
| `workspace` | `string` | no | Override workspace (default: from each file's engram_workspace) |
| `confirm` | `boolean` | no | Apply writes. When false (default), dry-run review only Default: `false`. |
| `force_version` | `boolean` | no | Bypass version conflict checks Default: `false`. |

### `memory_block_create`

Create a named, token-bounded memory block (Letta/MemGPT-style self-editing context slot).

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `name` | `string` | yes | Unique name for the memory block |
| `content` | `string` | no | Initial content of the block (default: empty string) |
| `max_tokens` | `integer` | no | Maximum token capacity for the block (default: 4096) |

### `memory_block_get`

Retrieve a memory block by name.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `name` | `string` | yes | Name of the memory block to retrieve |

### `memory_block_edit`

Update the content of an existing memory block, incrementing its version and recording the reason.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `content`, `name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `name` | `string` | yes | Name of the memory block to edit |
| `content` | `string` | yes | New content for the block |
| `reason` | `string` | no | Human-readable reason for this edit (optional) |

### `memory_block_list`

List all memory blocks with their names, versions, and token usage.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_block_archive`

Permanently delete a memory block and return its final content before deletion. Destructive and irreversible.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `name` | `string` | yes | Name of the memory block to archive and delete |

### `memory_block_history`

Return the edit history for a named memory block.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `name` | `string` | yes | Name of the memory block |
| `limit` | `integer` | no | Maximum number of history entries to return (default: 20) |

### `memory_cache_stats`

Return hit/miss statistics and entry count for the in-memory semantic search cache.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_cache_clear`

Evict all entries from the semantic search cache. Mutates in-memory cache state.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_compress_for_context`

Pack a set of memories into a token budget for LLM context, returning compressed entries and diagnostics about skipped memories.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `ids`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `ids` | `array` | yes | Memory IDs to compress and pack (alias: memory_ids). Items: `integer`. |
| `memory_ids` | `array` | no | Alias for ids. Items: `integer`. |
| `token_budget` | `integer` | no | Maximum token budget for the packed context (default: 4096). |

### `memory_embedding_migrate`

Re-embed all memories using the active embedding model; use dry_run to count affected memories without writing.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `dry_run` | `boolean` | no | If true, count memories to migrate without re-embedding them (default: false). |
| `target_model` | `string` | no | Target embedding model name to record in embedding_model column. Defaults to the active embedder's model name. |

### `memory_embedding_providers`

List the active embedding provider including model name and vector dimensions.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `temporal_add_edge`

Add a bi-temporal validity edge between two memories in the knowledge graph.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `from_id`, `relation`, `to_id`, `valid_from`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `from_id` | `integer` | yes | Source memory ID. |
| `to_id` | `integer` | yes | Target memory ID. |
| `relation` | `string` | yes | Semantic label for the edge (e.g. "works_at"). |
| `valid_from` | `string` | yes | RFC3339 timestamp marking the start of edge validity. |
| `properties` | `object` | no | Arbitrary JSON metadata to attach to the edge. |
| `confidence` | `number` | no | Edge confidence score between 0.0 and 1.0 (default: 1.0). |
| `source` | `string` | no | Provenance string identifying where this edge originates. |
| `scope_path` | `string` | no | Optional scope path to associate with this edge. |

### `temporal_contradictions`

Detect overlapping or contradictory edge pairs in the temporal knowledge graph.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `temporal_diff`

Compute the set of added, removed, and changed edges between two RFC3339 timestamps in the temporal graph.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `t1`, `t2`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `t1` | `string` | yes | Earlier RFC3339 timestamp (snapshot baseline). |
| `t2` | `string` | yes | Later RFC3339 timestamp (snapshot target). |
| `scope_path` | `string` | no | Optional scope path to restrict the diff. |

### `temporal_snapshot`

Return all currently-valid temporal graph edges as of a given RFC3339 timestamp.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `timestamp`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `timestamp` | `string` | yes | RFC3339 point-in-time for the snapshot. |
| `scope_path` | `string` | no | Optional scope path to restrict the snapshot. |

### `temporal_timeline`

Return the full edge history between two memory IDs, ordered chronologically.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `from_id`, `to_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `from_id` | `integer` | yes | Source memory ID. |
| `to_id` | `integer` | yes | Target memory ID. |
| `scope_path` | `string` | no | Optional scope path to restrict the timeline. |

### `memory_enrichment_timeline`

List all enrichment events for a specific memory (lifecycle transitions, consolidation, compression, etc.). Shows what automated operations affected this memory and why.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory whose enrichment history to retrieve. |
| `event_type` | `string` | no | Filter to a specific event type (e.g. "consolidation", "lifecycle_transition"). |
| `include_dry_runs` | `boolean` | no | Include events that were executed in dry-run mode (default: true). |
| `include_snapshots` | `boolean` | no | Include snapshot events (default: true). |
| `limit` | `integer` | no | Maximum number of events to return (default: 20, max: 100). |

### `memory_enrichment_audit`

Query enrichment events globally with filters (status, event_type, agent_id, operation_id, workspace, time range). Use for compliance audit and batch tracing.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `event_type` | `string` | no | Filter by event type (e.g. "consolidation", "lifecycle_transition", "compression"). |
| `triggered_by` | `string` | no | Filter by the tool name that triggered the event. |
| `agent_id` | `string` | no | Filter by the agent ID that triggered the event. |
| `status` | `string` | no | Filter by event outcome status. Allowed: `completed`, `failed`, `skipped`. |
| `workspace` | `string` | no | Filter to a specific workspace. |
| `operation_id` | `string` | no | Filter by a specific operation ID (exact match). |
| `memory_id` | `integer` | no | Filter to events that reference a specific memory. |
| `version_id` | `integer` | no | Filter to events that reference a specific memory version. |
| `dry_run` | `boolean` | no | Filter by dry-run flag (true = only dry-run events, false = only real events). |
| `since` | `string` | no | ISO-8601 timestamp: return events created at or after this time. |
| `until` | `string` | no | ISO-8601 timestamp: return events created at or before this time. |
| `order` | `string` | no | Sort order by creation time: "desc" (newest first, default) or "asc". Allowed: `desc`, `asc`. |
| `limit` | `integer` | no | Maximum number of events to return (default: 50, max: 200). |

### `memory_search`

Search memories using hybrid search (keyword + semantic). Automatically selects optimal strategy with optional reranking. Supports workspace isolation, tier filtering, and advanced filters.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query |
| `limit` | `integer` | no | Default: `10`. |
| `min_score` | `number` | no | Default: `0.1`. |
| `tags` | `array` | no | Items: `string`. |
| `memory_type` | `string` | no | Filter by memory type (preferred field; alias: type) |
| `type` | `string` | no | Deprecated alias for memory_type |
| `workspace` | `string` | no | Filter by single workspace |
| `workspaces` | `array` | no | Filter by multiple workspaces Items: `string`. |
| `tier` | `string` | no | Filter by memory tier Allowed: `permanent`, `daily`. |
| `include_transcripts` | `boolean` | no | Include transcript chunk memories (excluded by default) Default: `false`. |
| `strategy` | `string` | no | Force specific strategy (auto selects based on query; keyword/semantic are aliases for keyword_only/semantic_only) Allowed: `auto`, `keyword`, `keyword_only`, `semantic`, `semantic_only`, `hybrid`. |
| `explain` | `boolean` | no | Include match explanations Default: `false`. |
| `rerank` | `boolean` | no | Apply reranking to improve result quality Default: `true`. |
| `rerank_strategy` | `string` | no | Reranking strategy to use Default: `heuristic`. Allowed: `none`, `heuristic`, `multi_signal`. |
| `policy_rerank` | `boolean` | no | Apply memory policy retrieval_priority as an opt-in rerank layer after hybrid search. Default: `false`. |
| `policy_explain` | `boolean` | no | Include policy score and reason for each reranked result when policy_rerank is true. Default: `false`. |
| `filter` | `object` | no | Advanced filter with AND/OR logic. Supports workspace, tier, and metadata fields. Example: {"AND": [{"workspace": {"eq": "my-project"}}, {"importance": {"gte": 0.5}}]} |
| `global` | `boolean` | no | Search across all workspaces (default: false). When true, ignores any workspace filter and returns results from all workspaces with a workspace field in each result. Default: `false`. |

### `memory_smart_retrieve`

Intent-aware unified retrieval. Classifies the query (lookup, exploration, context, path) and dispatches to the right combination of internal retrievers, then merges and dedupes results. Returns audit fields `intents_used` and `strategies_called`.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Natural-language query |
| `limit` | `integer` | no | Default: `10`. Minimum: `1`. Maximum: `100`. |
| `workspace` | `string` | no | Optional workspace filter |
| `force_intents` | `array` | no | Override the classifier (for testing/debugging) Items: `string`. |

### `memory_digest`

Build an actionable, source-linked digest for a topic by orchestrating existing read-only retrieval, graph, and operational-context tools. Returns summaries, source memory IDs, relationships, next actions, provenance, and warnings without mutating memory or invoking an LLM.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `topic`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `topic` | `string` | yes | Topic, task, decision, or question to summarize from memory |
| `workspace` | `string` | no | Optional workspace scope |
| `mode` | `string` | no | Digest depth and section size Default: `standard`. Allowed: `brief`, `standard`, `deep`. |
| `limit` | `integer` | no | Maximum source memories to inspect Default: `12`. Minimum: `1`. Maximum: `50`. |
| `related_depth` | `integer` | no | How many relationship hops to include from selected source memories Default: `1`. Minimum: `0`. Maximum: `2`. |
| `total_budget` | `integer` | no | Token budget passed to context assembly for accounting only; raw prompt content is not returned Default: `4096`. Minimum: `512`. Maximum: `12000`. |
| `include_types` | `array` | no | Optional memory_type allowlist for digest source memories Items: `string`. |
| `timeframe` | `string` | no | Time window for context assembly Default: `all`. Allowed: `1h`, `24h`, `7d`, `30d`, `all`. |
| `include_graph` | `boolean` | no | Include cross-reference relationships for source memories Default: `true`. |
| `include_operational_context` | `boolean` | no | Include compact Operational Context bundle sections Default: `true`. |
| `include_next_actions` | `boolean` | no | Include extractive next-action suggestions grounded in source IDs Default: `true`. |
| `current_git_branch` | `string` | no | Current branch used for Operational Context staleness warnings |
| `current_commit_hash` | `string` | no | Current commit used for Operational Context staleness warnings |

### `memory_council`

Run a question through an llm-council instance (Karpathy council orchestration) and return consolidated stage outputs and final answer. Optionally persist a checkpoint memory.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `prompt`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `prompt` | `string` | yes | Prompt to send to the council |
| `conversation_id` | `string` | no | Optional existing conversation ID to continue |
| `council_url` | `string` | no | Council HTTP base URL Default: `http://127.0.0.1:8001`. |
| `timeout_seconds` | `integer` | no | Request timeout in seconds (1-300) Default: `90`. Minimum: `1`. Maximum: `300`. |
| `include_raw_stages` | `boolean` | no | Whether to include raw stage payloads Default: `true`. |
| `persist` | `boolean` | no | Persist final answer as checkpoint memory Default: `false`. |
| `workspace` | `string` | no | Target workspace when persist=true Default: `default`. |
| `memory_tags` | `array` | no | Extra tags to include when persist=true (default tags: llm-council, consensus) Items: `string`. |

### `memory_search_suggest`

Get search suggestions and typo corrections

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | No description. |

### `memory_find_duplicates`

Find potential duplicate memories

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `threshold` | `number` | no | Default: `0.9`. |

### `memory_find_semantic_duplicates`

Find semantically similar memories using embedding cosine similarity (LLM-powered dedup). Goes beyond hash/n-gram to detect paraphrased content.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `threshold` | `number` | no | Cosine similarity threshold (0.92 = very similar) Default: `0.92`. |
| `workspace` | `string` | no | Filter by workspace (optional) |
| `limit` | `integer` | no | Maximum duplicate pairs to return Default: `50`. |

### `langfuse_connect`

Configure Langfuse connection for observability integration. Stores config in metadata.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `public_key` | `string` | no | Langfuse public key (or use LANGFUSE_PUBLIC_KEY env var) |
| `secret_key` | `string` | no | Langfuse secret key (or use LANGFUSE_SECRET_KEY env var) |
| `base_url` | `string` | no | Langfuse API base URL Default: `https://cloud.langfuse.com`. |

### `langfuse_sync`

Start background sync from Langfuse traces to memories. Returns task_id for status checking.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `since` | `string` | no | Sync traces since this timestamp (default: 24h ago) Format: `date-time`. |
| `limit` | `integer` | no | Maximum traces to sync Default: `100`. |
| `workspace` | `string` | no | Workspace to create memories in |
| `dry_run` | `boolean` | no | Preview what would be synced without creating memories Default: `false`. |

### `langfuse_sync_status`

Check the status of a Langfuse sync task.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `task_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `task_id` | `string` | yes | Task ID returned from langfuse_sync |

### `langfuse_extract_patterns`

Extract patterns from Langfuse traces without saving. Preview mode for pattern discovery.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `since` | `string` | no | Analyze traces since this timestamp Format: `date-time`. |
| `limit` | `integer` | no | Maximum traces to analyze Default: `50`. |
| `min_confidence` | `number` | no | Minimum confidence for patterns Default: `0.7`. |

### `search_cache_feedback`

Report feedback on search results quality. Helps tune the adaptive cache threshold.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `positive`, `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | The search query |
| `positive` | `boolean` | yes | True if results were helpful, false otherwise |
| `workspace` | `string` | no | Workspace filter used (if any) |

### `search_cache_stats`

Get search result cache statistics including hit rate, entry count, and current threshold.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `search_cache_clear`

Clear the search result cache. Useful after bulk operations.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Only clear cache for this workspace (optional) |

### `memory_search_by_identity`

Search memories by identity (person, entity, or alias). Finds all mentions of a specific identity across memories.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `identity`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `identity` | `string` | yes | Identity name or alias to search for |
| `workspace` | `string` | no | Optional: limit search to specific workspace |
| `limit` | `integer` | no | Maximum results to return Default: `50`. |

### `memory_session_search`

Search within session transcript chunks. Useful for finding content from past conversations.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query |
| `session_id` | `string` | no | Optional: limit to specific session |
| `workspace` | `string` | no | Optional: limit to specific workspace |
| `limit` | `integer` | no | Maximum results to return Default: `20`. |

### `meilisearch_search`

Search memories using Meilisearch (typo-tolerant, fast full-text). Requires Meilisearch to be configured. Falls back to hybrid search if unavailable.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query text |
| `limit` | `integer` | no | Maximum results to return Default: `20`. |
| `offset` | `integer` | no | Number of results to skip Default: `0`. |
| `workspace` | `string` | no | Filter by workspace |
| `tags` | `array` | no | Filter by tags (AND logic) Items: `string`. |
| `memory_type` | `string` | no | Filter by memory type |

### `meilisearch_reindex`

Trigger a full re-sync from SQLite to Meilisearch. Use after bulk imports or if the index is out of sync.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `meilisearch_status`

Get Meilisearch index status including document count, indexing state, and health.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `meilisearch_config`

Show current Meilisearch configuration (URL, sync interval, enabled status).

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_search_compact`

Token-efficient search returning only id, title (first line, max 80 chars), created_at, and tags. Use memory_expand to get full content for specific IDs.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query |
| `limit` | `integer` | no | Max results (default: 10) |
| `workspace` | `string` | no | Filter to workspace |
| `global` | `boolean` | no | Search across all workspaces (default: false). When true, ignores any workspace filter and includes a workspace field in each result. Default: `false`. |

### `memory_expand`

Fetch full memory content for specific IDs. Used after memory_search_compact to get full content only for memories you need.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `ids`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `ids` | `array` | yes | Memory IDs to expand Items: `integer`. |

### `memory_detect_updates`

Given new content, identify existing memories in a workspace that may be stale or in need of an update.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `content`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | New content to compare against stored memories. |
| `workspace` | `string` | no | Workspace to search for update candidates (default: "default"). |

### `memory_explain_search`

Explain how each result in a scored search batch was ranked, breaking down bm25, vector, fuzzy, recency, importance, and optional rerank contributions.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `results`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `results` | `array` | yes | Array of scored search result objects to explain. Items: `object`. |
| `reranking_active` | `boolean` | no | Whether cross-encoder reranking was active for this result set (default: false). |
| `rrf_k` | `integer` | no | RRF k constant used during retrieval (default: 60). |

### `memory_suggest_acquisitions`

Analyse knowledge gaps in a workspace and suggest new memories to create.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to analyse (default: "default"). |
| `limit` | `integer` | no | Maximum number of suggestions to return (default: 10). |

### `memory_link`

Create a cross-reference between two memories

- Tier: `essential`
- Annotations: mutating (no MCP hints)
- Required inputs: `from_id`, `to_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `from_id` | `integer` | yes | No description. |
| `to_id` | `integer` | yes | No description. |
| `edge_type` | `string` | no | Default: `related_to`. Allowed: `related_to`, `supersedes`, `contradicts`, `implements`, `extends`, `references`, `derived_from`, `depends_on`, `blocks`, `follows_up`. |
| `strength` | `number` | no | Relationship strength Minimum: `0`. Maximum: `1`. |
| `source_context` | `string` | no | Why this link exists |
| `pinned` | `boolean` | no | Exempt from confidence decay Default: `false`. |

### `memory_unlink`

Remove a cross-reference

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `from_id`, `to_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `from_id` | `integer` | yes | No description. |
| `to_id` | `integer` | yes | No description. |
| `edge_type` | `string` | no | Default: `related_to`. |

### `memory_related`

Get memories related to a given memory (depth>1 or include_entities returns traversal result)

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Starting memory ID |
| `depth` | `integer` | no | Traversal depth (1 = direct relations only) Default: `1`. |
| `include_entities` | `boolean` | no | Include connections through shared entities Default: `false`. |
| `edge_type` | `string` | no | Filter by edge type |
| `include_decayed` | `boolean` | no | Default: `false`. |

### `memory_extract_entities`

Extract named entities (people, organizations, projects, concepts) from a memory and store them

- Tier: `standard`
- Annotations: idempotentHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to extract entities from |

### `memory_get_entities`

Get all entities linked to a memory

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |

### `memory_search_entities`

Search for entities by name prefix

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query (prefix match) |
| `entity_type` | `string` | no | Filter by entity type (person, organization, project, concept, etc.) |
| `limit` | `integer` | no | Default: `20`. |

### `memory_entity_stats`

Get statistics about extracted entities

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_traverse`

Traverse the knowledge graph from a starting memory with full control over traversal options

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Starting memory ID |
| `depth` | `integer` | no | Maximum traversal depth Default: `2`. |
| `direction` | `string` | no | Default: `both`. Allowed: `outgoing`, `incoming`, `both`. |
| `edge_types` | `array` | no | Filter by edge types (related_to, depends_on, etc.) Items: `string`. |
| `min_score` | `number` | no | Minimum edge score threshold Default: `0`. |
| `min_confidence` | `number` | no | Minimum confidence threshold Default: `0`. |
| `limit_per_hop` | `integer` | no | Max results per hop Default: `50`. |
| `include_entities` | `boolean` | no | Include entity-based connections Default: `true`. |

### `memory_find_path`

Find the shortest path between two memories in the knowledge graph

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `from_id`, `to_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `from_id` | `integer` | yes | Starting memory ID |
| `to_id` | `integer` | yes | Target memory ID |
| `max_depth` | `integer` | no | Maximum path length to search Default: `5`. |

### `memory_graph_path`

Finds how two entities are connected in the knowledge graph via DuckDB OLAP engine. Discovers hidden relationships across multiple hops using recursive path-finding.

- Tier: `advanced`
- Annotations: readOnlyHint, idempotentHint
- Required inputs: `scope`, `source_id`, `target_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `scope` | `string` | yes | Tenant scope prefix, e.g., 'global/org/user' |
| `source_id` | `integer` | yes | Starting node ID |
| `target_id` | `integer` | yes | Target node ID |
| `max_depth` | `integer` | no | Maximum hops to traverse (default: 4, max: 10) Default: `4`. |

### `memory_temporal_snapshot`

Retrieves the exact facts and relationships that were true at a specific historical point in time. Uses DuckDB OLAP engine for fast columnar scans over temporal edges.

- Tier: `advanced`
- Annotations: readOnlyHint, idempotentHint
- Required inputs: `scope`, `timestamp`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `scope` | `string` | yes | Tenant scope prefix |
| `timestamp` | `string` | yes | ISO-8601 timestamp for the point-in-time query |

### `memory_scope_snapshot`

Compares the knowledge graph between two timestamps, showing what relationships were added, removed, or changed. Uses DuckDB OLAP engine for efficient temporal diff.

- Tier: `advanced`
- Annotations: readOnlyHint, idempotentHint
- Required inputs: `from_timestamp`, `scope`, `to_timestamp`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `scope` | `string` | yes | Tenant scope prefix |
| `from_timestamp` | `string` | yes | Start of comparison window (ISO-8601) |
| `to_timestamp` | `string` | yes | End of comparison window (ISO-8601) |

### `memory_auto_link`

Run semantic and temporal auto-linker on a workspace, creating crossref edges in the database. Mutates the database.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to auto-link (default: all workspaces) |
| `similarity_threshold` | `number` | no | Minimum cosine similarity to create a semantic link (default: 0.75) |
| `time_window_minutes` | `integer` | no | Time window in minutes for temporal linking (default: 30) |

### `memory_auto_link_stats`

Return aggregate statistics about auto-generated semantic and temporal links.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_cluster`

Run Louvain community detection on the memory graph and return detected clusters with modularity score.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `min_cluster_size` | `integer` | no | Minimum number of members for a cluster to be reported (default: 2). |
| `resolution` | `number` | no | Louvain resolution parameter controlling cluster granularity (default: 1.0). |
| `link_types` | `array` | no | Restrict clustering to these edge/link types. Omit to use all link types. Items: `string`. |

### `memory_coactivation_report`

Return coactivation graph statistics including edge count, average strength, and strongest co-occurring memory pairs.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_fact_graph`

Return all stored subject-predicate-object facts for a given subject entity.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `subject`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `subject` | `string` | yes | Entity name to look up in the facts table. |

### `memory_garden`

Run full autonomous garden maintenance on a workspace: prunes stale memories, merges duplicates, archives cold entries, and compresses verbose content.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to garden (default: "default"). |

### `memory_garden_preview`

Dry-run garden maintenance: reports what would be pruned, merged, archived, or compressed without making any changes.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to preview gardening for (default: "default"). |

### `memory_get_cluster`

Return the Louvain community cluster that contains a specific memory, including its cluster ID, size, and member IDs.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory whose cluster to look up. |

### `memory_knowledge_stats`

Return aggregate statistics over the knowledge-graph facts table: total facts, unique subjects/predicates/objects, and top entities.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_list_auto_links`

List auto-generated graph links (semantic or temporal) between memories, optionally filtered by link type.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `link_type` | `string` | no | Filter by link type: "semantic" or "temporal". Omit for all types. |
| `limit` | `integer` | no | Maximum number of links to return (default: 50). |

### `memory_list_clusters`

List all detected memory clusters from the persistent cluster table, optionally selecting the detection algorithm.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `algorithm` | `string` | no | Clustering algorithm to filter by (default: "louvain"). |

### `memory_list_facts`

List extracted subject-predicate-object facts, optionally scoped to a single source memory.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | no | Source memory ID to filter facts; omit to list facts from all memories. |
| `limit` | `integer` | no | Maximum number of facts to return (default: 100). |

### `memory_query_triplets`

SPARQL-like pattern query over the knowledge-graph facts table: match any combination of subject, predicate, and object (all optional, acts as wildcard when omitted).

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `subject` | `string` | no | Subject entity to match (wildcard if omitted). |
| `predicate` | `string` | no | Predicate/relation to match (wildcard if omitted). |
| `object` | `string` | no | Object value to match (wildcard if omitted). |

### `memory_reflect`

Generate a reflective synthesis over a set of memories at a configurable analytical depth (surface, analytical, or meta).

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `ids`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `ids` | `array` | yes | Array of memory IDs to reflect on (required, must be non-empty). Items: `integer`. |
| `depth` | `string` | no | Reflection depth: "surface" (default), "analytical", or "meta". |

### `memory_resolve_conflict`

Resolve a saved knowledge-graph conflict by ID using a chosen strategy, removing or retaining the conflicting edges accordingly.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `conflict_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `conflict_id` | `integer` | yes | ID of the conflict record to resolve (required). |
| `strategy` | `string` | no | Resolution strategy: "keep_newer" (default), "keep_higher_confidence", "merge", or "manual". |

### `memory_sentiment_analyze`

Analyze the sentiment of a single memory's content, returning a score, label (positive/neutral/negative), confidence, and keyword signals.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | ID of the memory to analyze (required). |

### `memory_sentiment_timeline`

Compute a chronological sentiment timeline over memories in a workspace within an optional time range.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to scan (default: "default"). |
| `from` | `string` | no | ISO-8601 start timestamp (default: epoch). |
| `to` | `string` | no | ISO-8601 end timestamp (default: far future). |
| `limit` | `integer` | no | Maximum number of timeline entries to return (default: 50). |

### `memory_sync_status`

Get cloud sync status

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_sync_media`

Sync local media assets (images, audio, video) to cloud S3/R2 storage. Uploads files from media_assets table that have not yet been synced. Returns a report of synced files. Requires both multimodal and cloud features.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `dry_run` | `boolean` | no | If true, report what would be synced without actually uploading Default: `false`. |

### `memory_events_poll`

Poll for memory events (create, update, delete, etc.) since a given point. Useful for syncing and monitoring.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `since_id` | `integer` | no | Return events after this event ID |
| `since_time` | `string` | no | Return events after this timestamp (RFC3339) Format: `date-time`. |
| `agent_id` | `string` | no | Filter events for specific agent |
| `limit` | `integer` | no | Maximum events to return Default: `100`. |

### `memory_events_clear`

Clear old events from the event log. Helps manage storage for long-running systems.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `before_id` | `integer` | no | Delete events before this ID |
| `before_time` | `string` | no | Delete events before this timestamp Format: `date-time`. |
| `keep_recent` | `integer` | no | Keep only the N most recent events |

### `sync_version`

Get the current sync version and metadata. Used to check if local data is up-to-date.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `sync_delta`

Get changes (delta) since a specific version. Returns created, updated, and deleted memories.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `since_version`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `since_version` | `integer` | yes | Version to get changes from |

### `sync_state`

Get or update sync state for a specific agent. Tracks what each agent has synced.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `agent_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | Agent identifier |
| `update_version` | `integer` | no | If provided, updates the agent's last synced version |

### `sync_cleanup`

Clean up old sync data (events, etc.) older than specified days.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `older_than_days` | `integer` | no | Delete sync data older than this many days Default: `30`. |

### `memory_share`

Share a memory with another agent. The target agent can poll for shared memories.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `from_agent`, `memory_id`, `to_agent`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of memory to share |
| `from_agent` | `string` | yes | Sender agent identifier |
| `to_agent` | `string` | yes | Recipient agent identifier |
| `message` | `string` | no | Optional message to include with share |

### `memory_shared_poll`

Poll for memories shared with this agent.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `agent_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | Agent identifier to check shares for |
| `include_acknowledged` | `boolean` | no | Include already acknowledged shares Default: `false`. |

### `memory_share_ack`

Acknowledge receipt of a shared memory.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `agent_id`, `share_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `share_id` | `integer` | yes | Share ID to acknowledge |
| `agent_id` | `string` | yes | Agent acknowledging the share |

### `memory_grant_access`

Grant an agent access to a scope path. Supports read, write, and admin permissions. Access also applies to all descendant scopes.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `agent_id`, `scope_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | Agent ID to grant access to |
| `scope_path` | `string` | yes | Scope path to grant access to (e.g. 'global/org:acme') |
| `permissions` | `string` | no | Permission level Default: `read`. Allowed: `read`, `write`, `admin`. |
| `granted_by` | `string` | no | Optional: ID of the granting agent |

### `memory_revoke_access`

Revoke an agent's access to a specific scope path.

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: `agent_id`, `scope_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | Agent ID to revoke access from |
| `scope_path` | `string` | yes | Scope path to revoke access from |

### `memory_list_grants`

List all scope access grants for a given agent.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `agent_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | Agent ID to list grants for |

### `memory_check_access`

Check whether an agent has a required permission level on a scope path (including ancestor grants).

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `agent_id`, `scope_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | Agent ID to check |
| `scope_path` | `string` | yes | Scope path to check access for |
| `permissions` | `string` | no | Required permission level Default: `read`. Allowed: `read`, `write`, `admin`. |

### `memory_search_by_image`

Search memories using an image as the query. Uses multimodal embeddings (CLIP-style) or falls back to describing the image via vision model and searching by description. Returns semantically similar memories — text or media — ranked by relevance.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `image_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `image_path` | `string` | yes | Path to the local image file to use as the search query |
| `limit` | `integer` | no | Maximum number of results to return Default: `10`. |
| `min_score` | `number` | no | Minimum similarity score (0.0-1.0) Minimum: `0`. Maximum: `1`. |
| `workspace` | `string` | no | Restrict search to a specific workspace |
| `strategy` | `string` | no | Embedding strategy: clip (requires CLIP embedder), description (vision model + text embedding), auto (use CLIP if available, else description) Default: `auto`. Allowed: `clip`, `description`, `auto`. |

### `memory_upload_image`

Upload an image file and attach it to a memory. The image will be stored locally and linked to the memory's metadata.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `file_path`, `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory to attach the image to |
| `file_path` | `string` | yes | Path to the image file to upload |
| `image_index` | `integer` | no | Index for ordering multiple images (0-based) Default: `0`. |
| `caption` | `string` | no | Optional caption for the image |

### `memory_migrate_images`

Migrate existing base64-encoded images in memories to file storage. Scans all memories and uploads any embedded data URIs to storage, replacing them with file references.

- Tier: `advanced`
- Annotations: idempotentHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `dry_run` | `boolean` | no | If true, only report what would be migrated without making changes Default: `false`. |

### `memory_capture_screenshot`

Capture a screenshot of the full screen or a specific application window and save it to a local file.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `app_name` | `string` | no | Name of the application window to capture; omit to capture the full screen |

### `memory_describe_image`

Describe the contents of an image file using the configured vision provider (requires VISION_PROVIDER env).

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `image_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `image_path` | `string` | yes | Absolute filesystem path to the image file (JPEG, PNG, WebP, etc.). |
| `prompt` | `string` | no | Optional custom prompt to guide the image description. |

### `memory_list_media`

List media assets stored in the media_assets table, optionally filtered by type (image, audio, video).

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `media_type` | `string` | no | Filter by media type: "image", "audio", or "video". Omit for all types. |
| `limit` | `integer` | no | Maximum number of assets to return (default: 50). |

### `memory_process_video`

Process a video file: extract metadata and keyframe descriptions via the configured vision provider, and create a memory record for the result.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `video_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `video_path` | `string` | yes | Absolute path to the video file to process. |

### `memory_transcribe_audio`

Transcribe an audio file to text using the configured audio transcription provider.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `audio_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `audio_path` | `string` | yes | Absolute or relative path to the audio file to transcribe. |

### `agent_register`

Register an AI agent with capabilities and namespace isolation. Upserts if agent_id already exists.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `agent_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | Unique identifier for the agent |
| `display_name` | `string` | no | Human-readable name (defaults to agent_id) |
| `capabilities` | `array` | no | List of capabilities (e.g., 'search', 'create', 'analyze') Items: `string`. |
| `namespaces` | `array` | no | Namespaces the agent operates in (default: ['default']) Items: `string`. |
| `metadata` | `object` | no | Additional metadata as key-value pairs |

### `agent_deregister`

Deregister an AI agent (soft delete — sets status to 'inactive').

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: `agent_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | ID of the agent to deregister |

### `agent_heartbeat`

Update an agent's heartbeat timestamp to indicate it is still alive.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `agent_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | ID of the agent sending heartbeat |

### `agent_list`

List registered agents, optionally filtered by status or namespace.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `status` | `string` | no | Filter by agent status Allowed: `active`, `inactive`. |
| `namespace` | `string` | no | Filter by namespace (returns agents that include this namespace) |

### `agent_get`

Get details of a specific registered agent by ID.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `agent_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | ID of the agent to retrieve |

### `agent_capabilities`

Update the capabilities list of a registered agent.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `agent_id`, `capabilities`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `agent_id` | `string` | yes | ID of the agent to update |
| `capabilities` | `array` | yes | New capabilities list (replaces existing) Items: `string`. |

### `snapshot_create`

Create a portable .egm snapshot of memories filtered by workspace, tags, date range, or importance. Optionally encrypt with AES-256-GCM or sign with Ed25519.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `output_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `output_path` | `string` | yes | File path for the .egm snapshot |
| `workspace` | `string` | no | Filter by workspace |
| `tags` | `array` | no | Filter by tags Items: `string`. |
| `importance_min` | `number` | no | Minimum importance score |
| `memory_types` | `array` | no | Filter by memory types Items: `string`. |
| `description` | `string` | no | Human-readable description |
| `creator` | `string` | no | Creator name |
| `encrypt_key` | `string` | no | Hex-encoded 32-byte AES key |
| `sign_key` | `string` | no | Hex-encoded 32-byte Ed25519 secret key |

### `snapshot_load`

Load a .egm snapshot into the memory store. Strategies: merge (skip duplicates), replace (clear workspace first), isolate (new workspace), dry_run (preview only).

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `path`, `strategy`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `path` | `string` | yes | Path to .egm file |
| `strategy` | `string` | yes | Load strategy Allowed: `merge`, `replace`, `isolate`, `dry_run`. |
| `target_workspace` | `string` | no | Target workspace (defaults to snapshot's workspace) |
| `decrypt_key` | `string` | no | Hex-encoded 32-byte AES key for encrypted snapshots |

### `snapshot_inspect`

Inspect a .egm snapshot without loading it. Returns manifest, file list, and size.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `path` | `string` | yes | Path to .egm file |

### `attestation_log`

Log a document ingestion with cryptographic attestation. Creates a chained record proving the document was processed.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `content`, `document_name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | Document content to attest |
| `document_name` | `string` | yes | Name of the document |
| `agent_id` | `string` | no | ID of the attesting agent |
| `memory_ids` | `array` | no | IDs of memories created from this document Items: `integer`. |
| `sign_key` | `string` | no | Hex-encoded 32-byte Ed25519 secret key |

### `attestation_verify`

Verify whether a document has been attested (ingested and recorded).

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `content`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | Document content to verify |

### `attestation_chain_verify`

Verify the integrity of the entire attestation chain. Returns valid, broken (with location), or empty.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `verifying_key` | `string` | no | Hex-encoded 32-byte Ed25519 verifying key. When provided, every record must carry a valid signature; missing or invalid signatures cause the chain to report as Broken. When omitted, only hash-chain integrity is verified. |

### `attestation_list`

List attestation records with optional filters. Supports JSON, CSV, and Merkle proof export formats.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `limit` | `integer` | no | Maximum records to return Default: `50`. |
| `offset` | `integer` | no | Number of records to skip Default: `0`. |
| `agent_id` | `string` | no | Filter by agent ID |
| `document_name` | `string` | no | Filter by document name |
| `export_format` | `string` | no | Export format Allowed: `json`, `csv`, `merkle_proof`. |

### `memory_agent_start`

Configure a tick-based memory agent for a workspace and return its initial configuration.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace the agent will operate on (default: "default") |
| `interval_secs` | `integer` | no | Desired check interval in seconds (default: 300) |

### `memory_agent_stop`

Stop a tick-based memory agent (no-op for stateless agents; resets client-side tracking).

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace whose agent should be stopped (default: "default") |

### `memory_agent_status`

Return current status and memory statistics for a workspace agent.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to report status for (default: "default") |

### `memory_agent_metrics`

Run one full agent cycle (prune/merge/archive) and return the actions taken and aggregate metrics. Mutates the database.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to run the agent cycle on (default: "default") |
| `max_actions` | `integer` | no | Maximum number of actions to take in this cycle (default: 10) |

### `harness_record`

Record a durable harness event (decision, handoff, failed_attempt, verification_result, risk, assumption, bug_reproduction, issue_update) with structured metadata for cross-session continuity. Use instead of memory_create when capturing work-state evidence rather than facts.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `kind`, `summary`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `kind` | `string` | yes | The harness event kind Allowed: `decision`, `handoff`, `failed_attempt`, `bug_reproduction`, `verification_result`, `risk`, `assumption`, `issue_update`. |
| `summary` | `string` | yes | Concise summary of the event (1-500 chars) Max length: `500`. |
| `details` | `string` | no | Optional additional context appended to the summary Max length: `8000`. |
| `source_paths` | `array` | no | File paths relevant to this event Items: `string`. |
| `command` | `string` | no | CLI/shell command that produced this evidence |
| `issue_number` | `integer` | no | Related GitHub issue number |
| `commit_sha` | `string` | no | Related git commit SHA |
| `evidence_refs` | `array` | no | Free-form references (URLs, paths, IDs) Items: `string`. |
| `importance` | `number` | no | Importance score (0-1) Default: `0.7`. Minimum: `0`. Maximum: `1`. |
| `workspace` | `string` | no | Workspace scope (default: 'default') |

### `harness_status`

Assemble current project state from harness memory records and optional git state. Returns current objective, active issues, recent decisions, known blockers, last verification, last handoff, and a suggested next action. Token-budget aware; degrades gracefully when git is unavailable.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace scope (default: 'default') |
| `max_records` | `integer` | no | Max recent harness records to include Default: `10`. Minimum: `1`. Maximum: `50`. |
| `token_budget` | `integer` | no | Approximate max tokens for the output (chars/4 heuristic) Default: `2000`. |
| `include_git` | `boolean` | no | Attempt to collect git branch/status/log state Default: `true`. |

### `harness_handoff`

Generate a structured handoff packet for next-agent continuity: current goal, files touched, decisions, tests run/not run, risks, blockers, and next steps. Optionally persists as a harness record. Does NOT claim completion unless verification_evidence is provided.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `current_goal`, `next_steps`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `current_goal` | `string` | yes | What the agent was working toward Max length: `300`. |
| `files_touched` | `array` | no | Paths modified this session Items: `string`. |
| `decisions_made` | `array` | no | Short decision summaries Items: `string`. |
| `tests_run` | `array` | no | Test commands/names that were run Items: `string`. |
| `tests_not_run` | `array` | no | Tests known to be missing or skipped Items: `string`. |
| `known_risks` | `array` | no | Open risks Items: `string`. |
| `blockers` | `array` | no | Things blocking progress Items: `string`. |
| `next_steps` | `array` | yes | Recommended actions for the next agent Items: `string`. Min items: `1`. |
| `issue_numbers` | `array` | no | Related GitHub issue numbers Items: `integer`. |
| `plan_doc_paths` | `array` | no | Paths to relevant plan docs Items: `string`. |
| `verification_evidence` | `string` | no | Evidence that work is complete (test count, command output summary) |
| `persist` | `boolean` | no | Persist the handoff as a harness record Default: `true`. |
| `workspace` | `string` | no | Workspace scope (default: 'default') |

### `harness_verify`

Record a verification command outcome with exit code, output summary, and optional evidence path/hash. Supports negative evidence (failures, skips with reason). Surfaces in harness_status as last_verification and feeds harness_handoff completion gating.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `command`, `exit_code`, `output_summary`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `command` | `string` | yes | The command that was run (e.g. 'cargo test --lib') Max length: `200`. |
| `exit_code` | `integer` | yes | Process exit code (0 = success) |
| `passed` | `boolean` | no | Explicit pass/fail; derived from exit_code == 0 if omitted |
| `output_summary` | `string` | yes | Concise summary (e.g. '873 tests passed, 0 failed') Max length: `500`. |
| `evidence_path` | `string` | no | Path to the full output file or log |
| `evidence_hash` | `string` | no | SHA-256 of the full output for integrity |
| `skipped_reason` | `string` | no | If skipped, why (negative evidence) |
| `issue_numbers` | `array` | no | Linked GitHub issues Items: `integer`. |
| `memory_ids` | `array` | no | Linked memory IDs Items: `integer`. |
| `importance` | `number` | no | Importance score (0-1) Default: `0.8`. Minimum: `0`. Maximum: `1`. |
| `workspace` | `string` | no | Workspace scope (default: 'default') |

### `memory_suggest_tags`

Suggest tags for a memory based on AI content analysis. Uses pattern matching, keyword extraction, and structure detection to suggest relevant tags with confidence scores.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | no | Memory ID to analyze (alternative to content) |
| `memory_id` | `integer` | no | Memory ID to analyze (alias for id) |
| `content` | `string` | no | Content to analyze (alternative to id/memory_id) |
| `type` | `string` | no | Memory type (used when providing content directly) Allowed: `note`, `todo`, `issue`, `decision`, `preference`, `learning`, `context`, `credential`. |
| `existing_tags` | `array` | no | Tags already on the memory (excluded from suggestions) Items: `string`. |
| `min_confidence` | `number` | no | Minimum confidence threshold for suggestions Default: `0.5`. Minimum: `0`. Maximum: `1`. |
| `max_tags` | `integer` | no | Maximum number of tags to suggest Default: `5`. |
| `enable_patterns` | `boolean` | no | Use pattern-based tagging Default: `true`. |
| `enable_keywords` | `boolean` | no | Use keyword-based tagging Default: `true`. |
| `enable_entities` | `boolean` | no | Use entity-based tagging Default: `true`. |
| `enable_type_tags` | `boolean` | no | Add tags based on memory type Default: `true`. |
| `keyword_mappings` | `object` | no | Custom keyword-to-tag mappings (e.g., {"ibvi": "project/ibvi"}) |

### `memory_auto_tag`

Automatically suggest and optionally apply tags to a memory. Analyzes content using AI heuristics and can merge suggested tags with existing ones.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to auto-tag |
| `memory_id` | `integer` | no | Memory ID (alias for id) |
| `apply` | `boolean` | no | If true, apply the suggested tags to the memory. If false, only return suggestions. Default: `false`. |
| `merge` | `boolean` | no | If true and apply=true, merge with existing tags. If false, replace existing tags. Default: `true`. |
| `min_confidence` | `number` | no | Minimum confidence threshold Default: `0.5`. Minimum: `0`. Maximum: `1`. |
| `max_tags` | `integer` | no | Maximum tags to suggest/apply Default: `5`. |
| `keyword_mappings` | `object` | no | Custom keyword-to-tag mappings |

### `session_context_create`

Create a new session context for tracking related memories during a conversation or task.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `name` | `string` | yes | Session name |
| `description` | `string` | no | Session description |
| `workspace` | `string` | no | Workspace for the session |
| `metadata` | `object` | no | Additional session metadata |

### `session_context_add_memory`

Add a memory to a session context with relevance score and role.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `memory_id`, `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID |
| `memory_id` | `integer` | yes | Memory ID to add |
| `relevance_score` | `number` | no | How relevant this memory is to the session Default: `1.0`. Minimum: `0`. Maximum: `1`. |
| `context_role` | `string` | no | Role of the memory in the session Default: `referenced`. Allowed: `referenced`, `created`, `updated`, `pinned`. |

### `session_context_remove_memory`

Remove a memory from a session context.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `memory_id`, `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID |
| `memory_id` | `integer` | yes | Memory ID to remove |

### `session_context_get`

Get a session context with its linked memories.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID |

### `session_context_list`

List all session contexts with optional filtering.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Filter by workspace |
| `active_only` | `boolean` | no | Only return active sessions Default: `false`. |
| `limit` | `integer` | no | Maximum sessions to return Default: `50`. |
| `offset` | `integer` | no | Offset for pagination Default: `0`. |

### `session_context_search`

Search memories within a specific session context.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `query`, `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID to search within |
| `query` | `string` | yes | Search query |
| `limit` | `integer` | no | Maximum results Default: `20`. |

### `session_context_update_summary`

Update the summary of a session context.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `session_id`, `summary`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID |
| `summary` | `string` | yes | New session summary |

### `session_context_end`

End a session context, marking it as inactive.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID to end |
| `summary` | `string` | no | Optional final summary |

### `session_context_export`

Export a session context with all its memories for archival or sharing.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session ID to export |
| `include_content` | `boolean` | no | Include full memory content Default: `true`. |
| `format` | `string` | no | Export format Default: `json`. Allowed: `json`, `markdown`. |

### `memory_get_public`

Get a memory with all <private>...</private> tagged sections removed. Safe for sharing in multi-agent contexts.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |

### `memory_get_injection_prompt`

Assembles the most relevant memories into a ready-to-inject system prompt block. Uses hybrid search to find relevant memories and formats them as markdown, respecting a token budget.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query to find relevant memories |
| `token_budget` | `integer` | no | Max tokens for output (default: 2000) |
| `workspace` | `string` | no | Filter to specific workspace |
| `include_types` | `array` | no | Filter by memory types Items: `string`. |

### `memory_observe_tool_use`

Store a tool observation as an episodic memory for session continuity. Automatically compresses large inputs/outputs.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `tool_input`, `tool_name`, `tool_output`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `tool_name` | `string` | yes | Name of the tool that was used |
| `tool_input` | `object` | yes | Tool input parameters |
| `tool_output` | `string` | yes | Tool output/result |
| `session_id` | `string` | no | Session identifier for grouping observations |
| `compress` | `boolean` | no | Compress to 200-char previews (default: true) |

### `memory_archive_tool_output`

Archives a tool's full raw output to memory and returns a compressed summary (~500 tokens) for use in the active context. Transforms O(N²) context growth to O(N) by keeping only summaries in the working context while preserving full outputs for on-demand retrieval.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `raw_output`, `tool_name`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `tool_name` | `string` | yes | Name of the tool whose output is being archived |
| `raw_output` | `string` | yes | Full raw output to archive |
| `session_id` | `string` | no | Session identifier for grouping archived outputs (default: 'unknown') |
| `compress_summary` | `boolean` | no | Whether to generate a compressed summary (default: true) |
| `summary_tokens` | `integer` | no | Max tokens for the compressed summary (default: 500) |

### `memory_get_archived_output`

Retrieves the full raw output for an archived tool observation by its archive ID. Use when you need the complete output that was previously compressed for context efficiency.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `archive_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `archive_id` | `integer` | yes | Archive ID returned by memory_archive_tool_output |

### `memory_get_working_memory`

Assembles all compressed tool observations for a session into a token-budgeted working memory block. Includes archive references for retrieving full outputs on demand. This is the core of the Endless Mode context management system.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session identifier to retrieve observations for |
| `token_budget` | `integer` | no | Max tokens for the working memory block (default: 4000) |
| `include_tool_names` | `array` | no | Whitelist of tool names to include (default: all) Items: `string`. |
| `since_minutes` | `integer` | no | Only include observations from the last N minutes (default: all time) |

### `session_land`

Generate a structured session handoff ('land the plane'). Creates a checkpoint memory with session summary, open items, recent decisions, and a bootstrap prompt for the next session. Call this at the end of every work session for seamless continuity.

- Tier: `essential`
- Annotations: mutating (no MCP hints)
- Required inputs: `session_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session identifier to hand off |
| `workspace` | `string` | no | Workspace scope (default: 'default') |
| `summary` | `string` | no | Summary of what was accomplished this session |
| `next_session_hints` | `array` | no | Hints for what should be done next session Items: `string`. |

### `memory_build_context`

Build a structured prompt context from relevant memories using hybrid search, with optional graph traversal depth, timeframe filtering, type filtering, and relationship graph inclusion. Inspired by Basic Memory's build_context.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query to retrieve relevant memories |
| `total_budget` | `integer` | no | Max tokens for the entire prompt (default: 4096) |
| `strategy` | `string` | no | Context assembly strategy Default: `greedy`. Allowed: `greedy`, `balanced`, `recency`. |
| `workspace` | `string` | no | Workspace to search in |
| `limit` | `integer` | no | Max memories to retrieve (default: 20) |
| `depth` | `integer` | no | Graph traversal depth: 1=search only, 2=search+1 hop of related memories, 3=search+2 hops Default: `1`. Minimum: `1`. Maximum: `3`. |
| `timeframe` | `string` | no | Time window for memory filtering Default: `all`. Allowed: `1h`, `24h`, `7d`, `30d`, `all`. |
| `include_types` | `array` | no | Only include these memory types (e.g., ['note', 'decision']) Items: `string`. |
| `include_graph` | `boolean` | no | Include entity relationship graph in response Default: `false`. |

### `context_record`

Record a scoped Operational Context event and optional derived summary. Redacts text before storage, requires provenance scope, keeps raw payload storage off, and supports RTK-compatible external summary metadata without dereferencing external pointers.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `event_type`, `session_id`, `source`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `source` | `string` | yes | Source system or adapter, for example codex, harness, rtk, github_actions |
| `source_version` | `string` | no | Optional source or adapter version |
| `repo_id` | `string` | no | Repository scope identifier, for example github:aiconnai/engram |
| `workspace_path_hash` | `string` | no | Workspace path hash scope |
| `workspace` | `string` | no | Alias for workspace_path_hash |
| `git_branch` | `string` | no | Branch observed when the event occurred |
| `worktree_name` | `string` | no | Worktree name observed when the event occurred |
| `commit_hash` | `string` | no | Commit observed when the event occurred |
| `session_id` | `string` | yes | Agent or user session scope |
| `task_id` | `string` | no | Task, issue, or ticket scope |
| `agent_id` | `string` | no | Agent identity |
| `event_type` | `string` | yes | Event family such as command, tool, decision_made, verification_run, verification_skipped, blocker_found, review_result, handoff_created |
| `command` | `string` | no | Command line or command name for command events |
| `command_name` | `string` | no | Alias/preferred command field |
| `tool` | `string` | no | Tool name for tool events |
| `tool_name` | `string` | no | Alias/preferred tool field |
| `cwd` | `string` | no | Working directory context |
| `exit_code` | `integer` | no | Command exit code; required for event_type=command |
| `summary` | `string` | no | Optional derived/lossy summary to store with provenance |
| `key_errors` | `array` | no | Important errors to index after redaction Items: `string`. |
| `touched_files` | `array` | no | Files inspected or changed by the event Items: `string`. |
| `reducer` | `object` | no | Optional reducer metadata: name, version, lossy, confidence, structured_facts, warnings, labels, tokens_raw_est, tokens_compact_est |
| `external_reducer` | `string` | no | External reducer name for RTK-compatible summaries |
| `raw_pointer` | `string` | no | External raw pointer recorded as metadata only; Engram does not dereference it |
| `external_unverified` | `boolean` | no | Mark external summary as unverified |
| `labels` | `array` | no | Additional labels; external records add derived/lossy/external_unverified conservatively Items: `string`. |
| `retention_policy` | `string` | no | Retention label; sensitive commands may be forced to ephemeral_sensitive by policy |
| `raw_artifact_id` | `string` | no | Optional existing artifact id pointer; raw payload content is not accepted by this tool |
| `metadata` | `object` | no | Additional metadata redacted recursively before storage |
| `started_at` | `string` | no | RFC3339 event start time; defaults to now |
| `finished_at` | `string` | no | RFC3339 event finish time |

### `context_record_artifact`

Record an Operational Context artifact pointer or explicitly retained redacted raw artifact. Pointer-only is the default; raw_content requires retain_raw=true and policy approval.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `kind`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `string` | no | Optional artifact id; generated when omitted |
| `source_event_id` | `integer` | no | Related context event id |
| `source` | `string` | no | Source system or adapter |
| `source_version` | `string` | no | Optional source or adapter version |
| `repo_id` | `string` | no | Repository scope identifier |
| `workspace_path_hash` | `string` | no | Workspace path hash scope |
| `workspace` | `string` | no | Alias for workspace_path_hash |
| `session_id` | `string` | no | Session scope for access policy |
| `task_id` | `string` | no | Task scope for access policy |
| `agent_id` | `string` | no | Agent scope for access policy |
| `kind` | `string` | yes | Artifact type, for example command_output_summary, raw_command_output, review_artifact, diff_reference, test_report, external_url |
| `label` | `string` | no | Human label |
| `uri` | `string` | no | External/source-of-truth pointer |
| `raw_pointer` | `string` | no | Alias pointer stored as uri/metadata only; not dereferenced |
| `media_type` | `string` | no | Media type for pointer or raw content |
| `raw_content` | `string` | no | Raw content to retain only when retain_raw=true and policy allows it |
| `content_sha256` | `string` | no | Optional digest for pointer-only artifacts; raw digests are recomputed after redaction |
| `byte_len` | `integer` | no | Optional source byte length for pointer-only artifacts |
| `retention_policy` | `string` | no | Retention label, default pointer_only or raw_retained |
| `access_policy` | `string` | no | Default: `same_session`. Allowed: `same_session`, `same_task`, `same_agent`, `repo`, `public`. |
| `retain_raw` | `boolean` | no | Must be true to persist raw_content Default: `false`. |
| `ttl_seconds` | `integer` | no | Optional raw retention TTL from now |
| `stale_after_seconds` | `integer` | no | Optional stale threshold from now |
| `metadata` | `object` | no | Additional metadata redacted recursively before storage |

### `context_get_artifact`

Explicitly retrieve retained Operational Context artifact content after access, retention, staleness, and redaction checks. Search and bundle tools return artifact pointers only; this tool requires an artifact_id and reason.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `artifact_id`, `reason`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `artifact_id` | `string` | yes | Artifact ID to retrieve; broad search queries are not accepted |
| `reason` | `string` | yes | Why raw or retained artifact content is needed |
| `requester_agent_id` | `string` | no | Agent identity used for same_agent access checks |
| `session_id` | `string` | no | Session scope used for same_session access checks |
| `task_id` | `string` | no | Task scope used for same_task access checks |
| `repo_id` | `string` | no | Repository scope used for repo access checks |
| `workspace_path_hash` | `string` | no | Workspace path hash scope used for repo/workspace access checks |
| `workspace` | `string` | no | Alias for workspace_path_hash |
| `max_bytes` | `integer` | no | Maximum raw bytes to return; response marks truncation explicitly Minimum: `1`. |
| `allow_stale` | `boolean` | no | Allow retrieval after stale_at has passed Default: `false`. |
| `require_redacted` | `boolean` | no | Require a redaction status that permits raw retrieval Default: `true`. |

### `context_search`

Search scoped Operational Context events and derived summaries. Searches event metadata, command/tool names, summaries, structured facts, failure signals, decisions, inspected/touched file metadata, and artifact pointers without returning raw artifact content.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Search query for operational events, summaries, decisions, failures, files, or artifact metadata |
| `repo_id` | `string` | no | Repository scope identifier, for example github:aiconnai/engram |
| `workspace_path_hash` | `string` | no | Workspace path hash scope |
| `workspace` | `string` | no | Alias for workspace_path_hash when clients only have a workspace scope value |
| `session_id` | `string` | no | Session scope filter |
| `task_id` | `string` | no | Task scope filter |
| `event_type` | `string` | no | Restrict results to one event type |
| `event_types` | `array` | no | Restrict results to these event types Items: `string`. |
| `event_type_filters` | `array` | no | Alias for event_types Items: `string`. |
| `failure_only` | `boolean` | no | Only include failures/errors inferred from exit_code or event_type Default: `false`. |
| `max_results` | `integer` | no | Maximum results to return Default: `25`. Minimum: `1`. Maximum: `200`. |
| `include_artifact_pointers` | `boolean` | no | Include artifact IDs/pointers only; raw artifact content is never returned Default: `false`. |
| `current_git_branch` | `string` | no | Current branch used to mark branch mismatch staleness |
| `current_commit_hash` | `string` | no | Current commit used to mark commit mismatch staleness |
| `stale_after_days` | `integer` | no | Age threshold for stale warnings Default: `7`. |

### `context_build_bundle`

Build a compact agent-ready Operational Context bundle for resuming work. Includes relevant failures, inferred unresolved blockers, recent decisions, commands already run, inspected/touched files, staleness warnings, and optional artifact pointers with provenance for every item. Does not include raw artifact content.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | no | Optional query describing the work to resume |
| `repo_id` | `string` | no | Repository scope identifier, for example github:aiconnai/engram |
| `workspace_path_hash` | `string` | no | Workspace path hash scope |
| `workspace` | `string` | no | Alias for workspace_path_hash when clients only have a workspace scope value |
| `session_id` | `string` | no | Session scope filter |
| `task_id` | `string` | no | Task scope filter |
| `max_results` | `integer` | no | Maximum operational context rows to inspect Default: `80`. Minimum: `1`. Maximum: `200`. |
| `section_limit` | `integer` | no | Maximum entries per bundle section Default: `12`. Minimum: `1`. Maximum: `50`. |
| `include_artifact_pointers` | `boolean` | no | Include artifact IDs/pointers only; raw artifact content is never returned Default: `false`. |
| `current_git_branch` | `string` | no | Current branch used to mark branch mismatch staleness |
| `current_commit_hash` | `string` | no | Current commit used to mark commit mismatch staleness |
| `stale_after_days` | `integer` | no | Age threshold for stale warnings Default: `7`. |

### `recent_activity`

Discover recently created or updated memories. Returns compact previews sorted by most recent activity. Useful for understanding what has changed recently.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Filter by workspace (omit for all workspaces) |
| `timeframe` | `string` | no | Time window for activity Default: `24h`. Allowed: `1h`, `24h`, `7d`, `30d`. |
| `limit` | `integer` | no | Max results to return Default: `20`. Minimum: `1`. Maximum: `100`. |
| `include_types` | `array` | no | Only include these memory types Items: `string`. |

### `discover_tools`

List available Engram tools by tier and category. Use this to progressively discover capabilities beyond the essential tool set. Returns tool names, descriptions, and tiers.

- Tier: `essential`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `tier` | `string` | no | Filter by tier: essential (~20 core tools), standard (~57 common tools), advanced (~104 specialized tools), all (everything) Default: `all`. Allowed: `essential`, `standard`, `advanced`, `all`. |
| `category` | `string` | no | Filter by category keyword (e.g., 'search', 'graph', 'session', 'identity', 'quality') |
| `search` | `string` | no | Search tool names and descriptions |

### `memory_prepare_context`

Prepare optimized context for LLM using RTK-inspired pipeline (filter, group, truncate). Reduces token usage by 70-95% through intelligent context preparation.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `query`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Query to prepare context for |
| `budget` | `integer` | no | Token budget for prepared context Default: `4000`. |
| `workspace` | `string` | no | Optional workspace filter |

### `memory_extract_facts`

Extract subject-predicate-object facts from a memory's content using rule-based NLP and persist them to the facts table.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory from which to extract and store facts. |

### `scope_get`

Return the current scope path and level for a given memory.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory whose scope to retrieve. |

### `scope_list`

List all distinct scope paths currently present in the database.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `scope_search`

Search for memories whose content matches a query within a given scope, including ancestor scopes.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `query`, `scope_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | Substring to search for within scoped memories. |
| `scope_path` | `string` | yes | Hierarchical scope path to search within (e.g. "global/org:acme/user:alice"). |

### `scope_set`

Assign or update the hierarchical scope of a memory.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `memory_id`, `scope_path`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory to re-scope. |
| `scope_path` | `string` | yes | Target scope path (e.g. "global/org:acme/user:alice"). |

### `scope_tree`

Return a hierarchical tree of all scopes in the database.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| _(none)_ |  | no | No input properties declared. |

### `memory_soft_trim`

Intelligently trim memory content while preserving context. Keeps the beginning (head) and end (tail) of content with an ellipsis in the middle. Useful for displaying long content in limited space while keeping important context from both ends.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to trim |
| `max_chars` | `integer` | no | Maximum characters for trimmed output Default: `500`. |
| `head_percent` | `integer` | no | Percentage of space for the head (0-100) Default: `60`. |
| `tail_percent` | `integer` | no | Percentage of space for the tail (0-100) Default: `30`. |
| `ellipsis` | `string` | no | Text to insert between head and tail Default: ` ... `. |
| `preserve_words` | `boolean` | no | Avoid breaking in the middle of words Default: `true`. |

### `memory_list_compact`

List memories with compact preview instead of full content. More efficient for browsing/listing UIs. Returns only essential fields and a truncated content preview with metadata about original content length.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `limit` | `integer` | no | Maximum memories to return Default: `20`. |
| `offset` | `integer` | no | Pagination offset Default: `0`. |
| `tags` | `array` | no | Filter by tags Items: `string`. |
| `memory_type` | `string` | no | Filter by memory type (preferred field; alias: type) |
| `type` | `string` | no | Deprecated alias for memory_type |
| `workspace` | `string` | no | Filter by workspace |
| `tier` | `string` | no | Filter by tier Allowed: `permanent`, `daily`. |
| `sort_by` | `string` | no | Default: `created_at`. Allowed: `created_at`, `updated_at`, `last_accessed_at`, `importance`, `access_count`. |
| `sort_order` | `string` | no | Default: `desc`. Allowed: `asc`, `desc`. |
| `preview_chars` | `integer` | no | Maximum characters for content preview Default: `100`. |

### `memory_content_stats`

Get content statistics for a memory (character count, word count, line count, sentence count, paragraph count)

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID |

### `memory_export`

Export all memories to a JSON-serializable format for backup or migration.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Optional: export only from specific workspace |
| `include_embeddings` | `boolean` | no | Include embedding vectors in export (larger file size) Default: `false`. |

### `memory_import`

Import memories from a previously exported JSON format.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `data`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `data` | `object` | yes | The exported data object |
| `skip_duplicates` | `boolean` | no | Skip memories with matching content hash Default: `true`. |

### `memory_create_section`

Create a section memory for organizing content hierarchically. Sections can have parent sections for nested organization.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `title`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `title` | `string` | yes | Section title |
| `content` | `string` | no | Section description or content |
| `parent_id` | `integer` | no | Optional parent section ID for nesting |
| `level` | `integer` | no | Heading level (1-6) Default: `1`. |
| `workspace` | `string` | no | Workspace for the section |

### `memory_checkpoint`

Create a checkpoint memory marking a significant point in a session. Useful for session resumption and context restoration.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `session_id`, `summary`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `session_id` | `string` | yes | Session identifier |
| `summary` | `string` | yes | Summary of session state at checkpoint |
| `context` | `object` | no | Additional context data to preserve |
| `workspace` | `string` | no | Workspace for the checkpoint |

### `memory_create_episodic`

Create an episodic memory representing an event with temporal context. Use for tracking when things happened and their duration.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `content`, `event_time`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | Description of the event |
| `event_time` | `string` | yes | ISO8601 timestamp when the event occurred Format: `date-time`. |
| `event_duration_seconds` | `integer` | no | Duration of the event in seconds |
| `tags` | `array` | no | Tags for categorization Items: `string`. |
| `metadata` | `object` | no | Additional metadata |
| `importance` | `number` | no | Importance score (0-1) Minimum: `0`. Maximum: `1`. |
| `workspace` | `string` | no | Workspace (default: 'default') |

### `memory_create_procedural`

Create a procedural memory representing a learned pattern or workflow. Tracks success/failure to measure effectiveness.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `content`, `trigger_pattern`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content` | `string` | yes | Description of the procedure/workflow |
| `trigger_pattern` | `string` | yes | Pattern that triggers this procedure (e.g., 'When user asks about auth') |
| `tags` | `array` | no | Tags for categorization Items: `string`. |
| `metadata` | `object` | no | Additional metadata |
| `importance` | `number` | no | Importance score (0-1) Minimum: `0`. Maximum: `1`. |
| `workspace` | `string` | no | Workspace (default: 'default') |

### `memory_get_timeline`

Query episodic memories by time range. Returns events ordered by event_time.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `start_time` | `string` | no | Start of time range (ISO8601) Format: `date-time`. |
| `end_time` | `string` | no | End of time range (ISO8601) Format: `date-time`. |
| `workspace` | `string` | no | Filter by workspace |
| `tags` | `array` | no | Filter by tags Items: `string`. |
| `limit` | `integer` | no | Maximum results to return Default: `50`. |

### `memory_get_procedures`

List procedural memories (learned patterns/workflows). Optionally filter by trigger pattern.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `trigger_pattern` | `string` | no | Filter by trigger pattern (partial match) |
| `workspace` | `string` | no | Filter by workspace |
| `min_success_rate` | `number` | no | Minimum success rate (successes / (successes + failures)) Minimum: `0`. Maximum: `1`. |
| `limit` | `integer` | no | Maximum results to return Default: `50`. |

### `memory_record_procedure_outcome`

Record a success or failure for a procedural memory. Increments the corresponding counter.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`, `success`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Procedural memory ID |
| `success` | `boolean` | yes | true = success, false = failure |

### `memory_boost`

Temporarily boost a memory's importance score. The boost can optionally decay over time.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to boost |
| `boost_amount` | `number` | no | Amount to increase importance (0-1) Default: `0.2`. |
| `duration_seconds` | `integer` | no | Optional: duration before boost decays (omit for permanent boost) |

### `memory_explain_utility`

Explain why a memory has its current utility score. Returns the full feedback history summary (useful vs. not-useful retrievals), how much temporal decay has been applied, and a plain-English narrative. Useful for debugging or auditing memory quality.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `memory_id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory to explain |

### `memory_summarize`

Create a summary of one or more memories. Returns a new Summary-type memory with summary_of_id set.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `memory_ids`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_ids` | `array` | yes | IDs of memories to summarize Items: `integer`. |
| `summary` | `string` | no | The summary text (provide this or let the system generate one) |
| `max_length` | `integer` | no | Maximum length for auto-generated summary Default: `500`. |
| `workspace` | `string` | no | Workspace for the summary memory |
| `tags` | `array` | no | Tags for the summary memory Items: `string`. |

### `memory_get_full`

Get the full/original content of a memory. If the memory is a Summary, returns the original content from summary_of_id.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to get full content for |

### `memory_auto_consolidate`

Enable, disable, configure, or inspect the automatic consolidation scheduler. Use action='enable'/'disable' to toggle it, 'set_interval' with interval_seconds to change the period (60–86400), or 'get_status' to inspect current settings.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: `action`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `action` | `string` | yes | Allowed: `enable`, `disable`, `set_interval`, `get_status`. |
| `interval_seconds` | `integer` | no | Minimum: `60`. Maximum: `86400`. |

### `memory_consolidate_batch`

Run one auto-consolidation pass over a workspace: detect duplicates, conflicts, and archive-eligible memories. Defaults to dry-run; returns a structured report of actions taken (or that would be taken).

- Tier: `advanced`
- Annotations: destructiveHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Default: `default`. |
| `dry_run` | `boolean` | no | Default: `true`. |
| `policy` | `object` | no | No description. |

### `memory_consolidation_history`

List recent auto-consolidation runs for a workspace (or all workspaces). Newest-first.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | No description. |
| `limit` | `integer` | no | Default: `20`. Minimum: `1`. Maximum: `1000`. |

### `memory_compress`

Apply rule-based semantic compression to a single memory and return the structured result with key entities and facts.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | ID of the memory to compress. |
| `target_ratio` | `number` | no | Target compression ratio as a fraction of original tokens (default: 0.1). |

### `memory_consolidate`

Run offline consolidation over a workspace, merging and archiving similar memories; use dry_run to preview without writing.

- Tier: `advanced`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace to consolidate (default: "default"). |
| `strategy` | `string` | no | Grouping strategy: "content_overlap" (default), "tag_similarity", or "temporal_proximity". |
| `dry_run` | `boolean` | no | If true, report what would be merged/archived without writing changes (default: false). |

### `memory_decompress`

Retrieve the original (uncompressed) content of a memory by ID.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | ID of the memory whose content to retrieve. |

### `memory_detect_conflicts`

Detect contradictory or conflicting facts in the knowledge graph; optionally persist detected conflicts for later resolution.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `save` | `boolean` | no | If true, persist detected conflicts to the conflicts table for later resolution (default: false). |

### `memory_feedback`

Record relevance feedback for a search result and update the memory's utility score; schedules low-utility memories for consolidation.

- Tier: `standard`
- Annotations: mutating (no MCP hints)
- Required inputs: `memory_id`, `query`, `signal`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `query` | `string` | yes | The search query that produced the result. |
| `memory_id` | `integer` | yes | ID of the memory being rated. |
| `signal` | `string` | yes | Feedback signal: "useful" (alias "helpful"), "irrelevant" (alias "not_helpful"), "outdated", or "conflict". |
| `rank_position` | `integer` | no | 0-based rank position of the result in the original result list (optional). |
| `original_score` | `number` | no | The final_score from the original search result (optional). |
| `workspace` | `string` | no | Workspace context for the feedback (default: "default"). |

### `memory_feedback_stats`

Return aggregated search-feedback statistics (thumbs-up/down counts, top-rated queries) for a workspace.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: none

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `workspace` | `string` | no | Workspace name to filter stats; omit for all workspaces. |

### `memory_synthesis`

Check semantic overlap between two content strings and produce a merged synthesis using the chosen strategy.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `content_a`, `content_b`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `content_a` | `string` | yes | First content string to synthesise. |
| `content_b` | `string` | yes | Second content string to synthesise. |
| `id_a` | `integer` | no | Optional memory ID associated with content_a (default: 0). |
| `strategy` | `string` | no | Synthesis strategy: "merge" (default), "replace", or "append". Allowed: `merge`, `replace`, `append`. |

### `memory_utility_score`

Compute the Q-value utility score for a memory based on its retrieval feedback history.

- Tier: `standard`
- Annotations: readOnlyHint
- Required inputs: `id`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `id` | `integer` | yes | Memory ID to score. |

### `memory_replay_at_time`

Replay one memory as it existed at a given RFC3339 timestamp and optionally return enrichment events affecting it up to that time.

- Tier: `advanced`
- Annotations: readOnlyHint
- Required inputs: `memory_id`, `timestamp`

| Input | Type | Required | Summary |
|-------|------|----------|---------|
| `memory_id` | `integer` | yes | ID of the memory to replay. |
| `timestamp` | `string` | yes | RFC3339 timestamp to replay state at. |
| `event_type` | `string` | no | Optional event type filter for replayed event list (e.g. "consolidation"). |
| `include_events` | `boolean` | no | Whether to include enrichment events in the response. Default: `true`. |
| `include_failed` | `boolean` | no | Whether to include failed enrichment events. Default: `false`. |
| `include_dry_runs` | `boolean` | no | Whether to include dry-run events. Default: `false`. |
| `event_limit` | `integer` | no | Max number of events to include in replay trail (default 50, max 200). |