briefcase-python 2.4.1

Python bindings for Briefcase AI
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
"""
Broker-dealer onboarding workflow for the Clearwater/Triton demo scenario.
"""

from __future__ import annotations

from dataclasses import asdict, dataclass
from datetime import datetime, timezone
import csv
import json
import re
from pathlib import Path
from statistics import mean
from typing import Any, Dict, List, Mapping, Optional, Tuple, Union

from briefcase.integrations.lakefs import ArtifactLineageClient

try:
    from pypdf import PdfReader  # type: ignore

    HAS_PYPDF = True
except ImportError:
    HAS_PYPDF = False


PathLike = Union[str, Path]

SOURCE_OBJECT_PATHS = {
    "A": "Clients/Active/Triton Capital Partners LP/Onboarding/"
    "Triton_Capital_NewAcct_Application_2026.pdf",
    "B": "Clients/Active/Triton Capital Partners LP/Onboarding/"
    "Triton_Capital_BeneficialOwnership_Cert.pdf",
    "C": "Clients/Active/Triton Capital Partners LP/Onboarding/"
    "Triton_Capital_Formation_Docs.pdf",
    "D": "Clients/Active/Triton Capital Partners LP/Onboarding/"
    "Triton_Capital_AuthorizedTraders.pdf",
    "E": "Clients/Active/Triton Capital Partners LP/Onboarding/Triton_Capital_W8BENE.pdf",
}

PLAYBOOK_OBJECT_PATH = "Playbooks/Institutional_Client_Onboarding_PB_v4.2.json"
ANALYTICS_OBJECT_PATH = "Analytics/Meridian_Submission_Pattern_History.csv"
ANALYSIS_SNAPSHOT_OBJECT_PATH = "Analytics/Triton_Capital_Rule_Evaluation_2026_02_03.json"
MEMO_ARTIFACT_OBJECT_PATH = "AuditArtifacts/DECISION_MEMO_Triton_Capital_Partners_LP.md"
RESULT_ARTIFACT_OBJECT_PATH = "AuditArtifacts/triton_onboarding_demo_result.json"


@dataclass
class WorkflowFlag:
    flag_id: str
    severity: str
    rules_triggered: List[str]
    source_documents: List[str]
    title: str
    what_found: str
    display_text: str
    significance: str
    source_commit_id: str
    analysis_commit_id: str
    source_object_uris: List[str]


class BrokerDealerOnboardingWorkflow:
    """
    Run the Clearwater Securities institutional onboarding workflow.

    The workflow versions source packets, analysis snapshots, and generated
    artifacts in lakeFS via `ArtifactLineageClient`.
    """

    def __init__(self, lineage_client: ArtifactLineageClient):
        self.lineage = lineage_client

    def run(
        self,
        document_paths: Mapping[str, PathLike],
        playbook_path: PathLike,
        output_dir: PathLike,
        meridian_history_csv: Optional[PathLike] = None,
        write_outputs: bool = True,
    ) -> Dict[str, Any]:
        """
        Execute workflow and return structured results.

        Args:
            document_paths: Mapping for keys A-E to onboarding document paths.
            playbook_path: Path to Institutional Client Onboarding playbook JSON.
            output_dir: Directory for generated artifacts.
            meridian_history_csv: Optional history CSV for pattern analysis.
            write_outputs: If true, write memo/result artifacts to disk.
        """
        docs = self._load_docs(document_paths)
        playbook = self._load_playbook(playbook_path)
        output_root = Path(output_dir)
        runtime_dir = output_root / ".runtime"
        runtime_dir.mkdir(parents=True, exist_ok=True)

        source_files = self._source_file_map(document_paths, playbook_path, meridian_history_csv)
        source_commit = self.lineage.version_files(
            files=source_files,
            message="Ingest Triton onboarding packet and Clearwater playbook",
            metadata={
                "workflow": "broker_dealer_onboarding",
                "client": "triton_capital_partners_lp",
                "event": "packet_ingest",
            },
        )

        analysis_snapshot_path = runtime_dir / "triton_onboarding_rule_evaluation_snapshot.json"
        analysis_snapshot_path.write_text(
            json.dumps(
                {
                    "generated_at_utc": datetime.now(timezone.utc).isoformat(),
                    "workflow": "Institutional Client Onboarding at a Broker-Dealer",
                    "scenario_client": "Triton Capital Partners LP",
                    "note": "Rule evaluation snapshot for audit lineage.",
                },
                indent=2,
            ),
            encoding="utf-8",
        )
        analysis_commit = self.lineage.version_files(
            files={ANALYSIS_SNAPSHOT_OBJECT_PATH: analysis_snapshot_path},
            message="Run playbook checks and persist rule-evaluation snapshot",
            metadata={
                "workflow": "broker_dealer_onboarding",
                "client": "triton_capital_partners_lp",
                "event": "rule_evaluation",
            },
        )

        flags, context = self._generate_flags(
            docs=docs,
            source_commit_id=source_commit.commit_id,
            analysis_commit_id=analysis_commit.commit_id,
            meridian_history_csv=meridian_history_csv,
        )

        memo_markdown = self._decision_memo_markdown(
            source_commit_id=source_commit.commit_id,
            analysis_commit_id=analysis_commit.commit_id,
            repository=source_commit.repository,
            branch=source_commit.branch,
        )

        result: Dict[str, Any] = {
            "generated_at_utc": datetime.now(timezone.utc).isoformat(),
            "scenario": {
                "name": "Briefcase Demo Build — Brett Taxin / White Bay Group",
                "workflow": "Institutional Client Onboarding at a Broker-Dealer",
                "broker_dealer": "Clearwater Securities LLC",
                "client": "Triton Capital Partners LP",
                "analyst": "Sarah Chen",
                "cco": "David Rourke",
            },
            "playbook": {
                "name": playbook.get("playbook_name"),
                "version": playbook.get("version"),
                "rules_loaded": self._count_rules(playbook),
            },
            "lakefs": {
                "mode": source_commit.mode,
                "repository": source_commit.repository,
                "branch": source_commit.branch,
                "source_packet_commit_id": source_commit.commit_id,
                "analysis_commit_id": analysis_commit.commit_id,
                "source_packet_paths": list(source_commit.files.keys()),
                "source_packet_uris": [
                    self.lineage.object_uri(path, source_commit.commit_id)
                    for path in source_commit.files.keys()
                ],
            },
            "rule_context": context,
            "flags": [asdict(flag) for flag in flags],
            "decision": {
                "risk_level": "ENHANCED",
                "approval_authority": "CCO (R-403)",
                "analyst_recommendation": (
                    "APPROVE with conditions (ongoing EDD monitoring for Triton Holdings Ltd.)"
                ),
                "cco_decision": "APPROVED",
                "account_activation_date": "2026-02-06",
            },
            "decision_memo_markdown": memo_markdown,
        }

        if write_outputs:
            output_root.mkdir(parents=True, exist_ok=True)
            memo_path = output_root / "DECISION_MEMO_Triton_Capital_Partners_LP.md"
            result_path = output_root / "triton_onboarding_demo_result.json"
            memo_path.write_text(memo_markdown, encoding="utf-8")
            result_path.write_text(json.dumps(result, indent=2), encoding="utf-8")

            output_commit = self.lineage.version_files(
                files={
                    MEMO_ARTIFACT_OBJECT_PATH: memo_path,
                    RESULT_ARTIFACT_OBJECT_PATH: result_path,
                },
                message="Publish decision memo artifacts for Triton onboarding",
                metadata={
                    "workflow": "broker_dealer_onboarding",
                    "client": "triton_capital_partners_lp",
                    "event": "artifact_publish",
                },
            )
            result["lakefs"]["memo_artifacts_commit_id"] = output_commit.commit_id
            result["lakefs"]["memo_artifact_uris"] = [
                self.lineage.object_uri(path, output_commit.commit_id)
                for path in output_commit.files.keys()
            ]
            result["artifacts"] = {
                "decision_memo_markdown": str(memo_path),
                "json_result": str(result_path),
            }
            result_path.write_text(json.dumps(result, indent=2), encoding="utf-8")

        return result

    def _load_docs(self, document_paths: Mapping[str, PathLike]) -> Dict[str, str]:
        missing = [key for key in ("A", "B", "C", "D", "E") if key not in document_paths]
        if missing:
            raise ValueError(f"document_paths missing required keys: {missing}")

        return {
            key: self._extract_text(Path(path))
            for key, path in document_paths.items()
            if key in {"A", "B", "C", "D", "E"}
        }

    def _load_playbook(self, playbook_path: PathLike) -> Dict[str, Any]:
        return json.loads(Path(playbook_path).read_text(encoding="utf-8"))

    @staticmethod
    def _count_rules(playbook: Dict[str, Any]) -> int:
        rule_sets = playbook.get("rule_sets", {})
        return sum(len(rules) for rules in rule_sets.values() if isinstance(rules, list))

    def _source_file_map(
        self,
        document_paths: Mapping[str, PathLike],
        playbook_path: PathLike,
        meridian_history_csv: Optional[PathLike],
    ) -> Dict[str, Path]:
        files: Dict[str, Path] = {
            SOURCE_OBJECT_PATHS[key]: Path(path) for key, path in document_paths.items()
            if key in SOURCE_OBJECT_PATHS
        }
        files[PLAYBOOK_OBJECT_PATH] = Path(playbook_path)
        if meridian_history_csv is not None:
            files[ANALYTICS_OBJECT_PATH] = Path(meridian_history_csv)
        return files

    def _generate_flags(
        self,
        docs: Dict[str, str],
        source_commit_id: str,
        analysis_commit_id: str,
        meridian_history_csv: Optional[PathLike],
    ) -> Tuple[List[WorkflowFlag], Dict[str, Any]]:
        bo_owners = self._parse_owners_from_bo_cert(docs["B"])
        formation_owners = self._parse_owners_from_formation_docs(docs["C"])

        # Known scenario fallback when text extraction is unavailable.
        if not bo_owners:
            bo_owners = {"Marcus Webb": 42.0, "Lena Okafor": 18.0}
        if not formation_owners:
            formation_owners = {
                "Marcus Webb": 42.0,
                "Lena Okafor": 18.0,
                "Triton Holdings Ltd.": 28.0,
                "Triton Capital GP LLC": 12.0,
            }

        formation_25_plus = {
            name: pct for name, pct in formation_owners.items() if pct >= 25 and "GP LLC" not in name
        }
        missing_from_bo = {
            name: pct for name, pct in formation_25_plus.items() if name not in bo_owners
        }

        bvi_trigger = "BVI" in docs["C"].upper() or "Triton Holdings Ltd." in docs["C"]
        missing_cco_signature = "[BLANK]" in docs["D"] or "CCO signature" in docs["D"].lower()

        meridian_stats = self._compute_meridian_pattern_stats(meridian_history_csv)
        faster_pct = int(round(meridian_stats["faster_processing_percentage"]))

        flags: List[WorkflowFlag] = []
        if "Triton Holdings Ltd." in missing_from_bo:
            source_objects = [SOURCE_OBJECT_PATHS["B"], SOURCE_OBJECT_PATHS["C"]]
            flags.append(
                WorkflowFlag(
                    flag_id="FLAG-1",
                    severity="CRITICAL",
                    rules_triggered=["R-101", "R-102", "R-303"],
                    source_documents=[
                        "Document B (Beneficial Ownership Certification)",
                        "Document C (Formation Docs)",
                    ],
                    title="Undisclosed Beneficial Owner",
                    what_found=(
                        "Formation docs show Triton Holdings Ltd. at 28% ownership and BVI "
                        "incorporation, but BO Certification omits that entity."
                    ),
                    display_text=(
                        "CRITICAL: Entity 'Triton Holdings Ltd.' holds 28% interest per LP "
                        "Agreement but is not listed on Beneficial Ownership Certification. "
                        "This exceeds the 25% CDD threshold. BVI incorporation triggers "
                        "enhanced due diligence and CCO approval under R-403."
                    ),
                    significance=(
                        "Cross-document discrepancy that appears complete at first glance "
                        "in Document B, but fails CDD once formation docs are considered."
                    ),
                    source_commit_id=source_commit_id,
                    analysis_commit_id=analysis_commit_id,
                    source_object_uris=self._uris_for(source_objects, source_commit_id),
                )
            )

        if missing_cco_signature:
            source_objects = [SOURCE_OBJECT_PATHS["D"]]
            flags.append(
                WorkflowFlag(
                    flag_id="FLAG-2",
                    severity="WARNING",
                    rules_triggered=["R-201"],
                    source_documents=["Document D (Authorized Trader List)"],
                    title="Incomplete Document (Missing Signature)",
                    what_found="Authorized trader list is missing Clearwater CCO signature.",
                    display_text=(
                        "WARNING: Authorized Trader List (Document D) is missing required CCO "
                        "signature. Rule R-201 requires dual authorization. Document is "
                        "incomplete and trading authority cannot be activated."
                    ),
                    significance="High-volume process miss that causes avoidable rework.",
                    source_commit_id=source_commit_id,
                    analysis_commit_id=analysis_commit_id,
                    source_object_uris=self._uris_for(source_objects, source_commit_id),
                )
            )

        source_objects = [ANALYTICS_OBJECT_PATH] if meridian_history_csv else []
        flags.append(
            WorkflowFlag(
                flag_id="FLAG-3",
                severity="INSIGHT",
                rules_triggered=["Pattern detection (Briefcase observability layer)"],
                source_documents=[
                    "Cross-file analysis of prior onboarding packets from Meridian Fund "
                    "Services Ltd."
                ],
                title="Pattern Anomaly from Fund Administrator",
                what_found=(
                    "Meridian packet language changed over time and correlates with faster "
                    "approvals and fewer escalations."
                ),
                display_text=(
                    "INSIGHT: Meridian Fund Services boilerplate language appears in recent "
                    f"submissions and correlates with {faster_pct}% faster processing. "
                    "This may indicate analyst reliance on administrator self-certification "
                    "in lieu of independent verification."
                ),
                significance=(
                    "Demonstrates input-layer observability beyond checklist controls."
                ),
                source_commit_id=source_commit_id,
                analysis_commit_id=analysis_commit_id,
                source_object_uris=self._uris_for(source_objects, source_commit_id),
            )
        )

        context = {
            "bo_cert_owners": bo_owners,
            "formation_owners": formation_owners,
            "owners_missing_from_bo_cert": missing_from_bo,
            "risk_rating": "ENHANCED" if bvi_trigger else "ELEVATED",
            "meridian_pattern_stats": meridian_stats,
            "source_commit_id": source_commit_id,
            "analysis_commit_id": analysis_commit_id,
        }
        return flags, context

    def _decision_memo_markdown(
        self,
        source_commit_id: str,
        analysis_commit_id: str,
        repository: str,
        branch: str,
    ) -> str:
        return (
            "# DECISION MEMO — Triton Capital Partners LP\n\n"
            "## Account Onboarding Review\n\n"
            "- Account: Triton Capital Partners LP — Prime Brokerage\n"
            "- Review initiated: February 3, 2026 at 9:14 AM EST\n"
            "- Analyst: Sarah Chen\n"
            "- Playbook applied: Clearwater Securities — Institutional Client "
            "Onboarding v4.2\n"
            "- Risk rating: ENHANCED (R-301 offshore domicile + R-303 BVI "
            "beneficial owner)\n"
            f"- lakeFS repository: {repository}\n"
            f"- lakeFS branch: {branch}\n"
            f"- Source packet commit ID: {source_commit_id}\n"
            f"- Rule-evaluation commit ID: {analysis_commit_id}\n\n"
            "## Approval\n\n"
            "- Analyst recommendation: APPROVE with conditions\n"
            "- CCO Decision: APPROVED\n"
            "- CCO Sign-off: David Rourke, 02/05/2026 at 3:08 PM EST\n"
            "- Account activation date: 02/06/2026\n\n"
            "Audit trail: Complete. All rule applications, flag resolutions, and "
            "approval actions logged with timestamps.\n"
        )

    def _compute_meridian_pattern_stats(self, csv_path: Optional[PathLike]) -> Dict[str, Any]:
        if csv_path is None or not Path(csv_path).exists():
            return {
                "total_packets": 17,
                "phrase_in_first_six": 0,
                "phrase_in_last_eleven": 9,
                "avg_hours_with_phrase": 14.4,
                "avg_hours_without_phrase": 24.0,
                "faster_processing_percentage": 40.0,
            }

        rows: List[Dict[str, str]] = []
        with Path(csv_path).open("r", encoding="utf-8") as handle:
            for row in csv.DictReader(handle):
                rows.append(row)

        first_six = rows[:6]
        last_eleven = rows[-11:]
        phrase_first = sum(1 for row in first_six if row.get("phrase_present") == "True")
        phrase_last = sum(1 for row in last_eleven if row.get("phrase_present") == "True")
        with_phrase = [
            float(row["processing_hours"])
            for row in rows
            if row.get("phrase_present") == "True" and row.get("processing_hours")
        ]
        without_phrase = [
            float(row["processing_hours"])
            for row in rows
            if row.get("phrase_present") == "False" and row.get("processing_hours")
        ]
        avg_with = mean(with_phrase) if with_phrase else 0.0
        avg_without = mean(without_phrase) if without_phrase else 0.0
        faster_pct = ((avg_without - avg_with) / avg_without * 100.0) if avg_without else 0.0

        return {
            "total_packets": len(rows),
            "phrase_in_first_six": phrase_first,
            "phrase_in_last_eleven": phrase_last,
            "avg_hours_with_phrase": round(avg_with, 2),
            "avg_hours_without_phrase": round(avg_without, 2),
            "faster_processing_percentage": round(faster_pct, 1),
        }

    def _uris_for(self, source_objects: List[str], source_commit_id: str) -> List[str]:
        return [self.lineage.object_uri(path, source_commit_id) for path in source_objects]

    @staticmethod
    def _extract_text(path: Path) -> str:
        if not path.exists():
            return ""

        if path.suffix.lower() == ".pdf" and HAS_PYPDF:
            reader = PdfReader(str(path))
            pages = [page.extract_text() or "" for page in reader.pages]
            return "\n".join(pages)

        try:
            return path.read_text(encoding="utf-8")
        except UnicodeDecodeError:
            return path.read_text(encoding="latin-1", errors="ignore")

    @staticmethod
    def _parse_owners_from_bo_cert(text: str) -> Dict[str, float]:
        owners: Dict[str, float] = {}
        pattern = re.compile(
            r"Beneficial Owner\s*#\d+:\s*(.*?),.*?(\d+)%\s*ownership",
            re.IGNORECASE,
        )
        for name, pct in pattern.findall(text):
            owners[name.strip()] = float(pct)
        return owners

    @staticmethod
    def _parse_owners_from_formation_docs(text: str) -> Dict[str, float]:
        owners: Dict[str, float] = {}
        pattern = re.compile(r"\(\d\)\s*(.*?)\s*[-—]\s*(\d+)%", re.IGNORECASE)
        for name, pct in pattern.findall(text):
            owners[name.strip()] = float(pct)
        return owners