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
/target/
**/target/
/Cargo.lock
**/*.rs.bk
.DS_Store
# Bench reporter outputs — regenerated by every captchaforge-bench run.
bench-results-*.csv
bench-results-*.html
bench-results-*.json
bench-results-*.md
bench-results-*.sarif
bench-results.json
# Trained model weights — large; ship via HuggingFace / Ollama / S3.
training/weights/
training/datasets/
# Python __pycache__ artefacts under tools/.
tools/__pycache__/
# ── Santh-internal artifacts (BEGIN — managed; do not edit between markers) ──
# Operator-private working notes, plans, audits, scoreboards, status
# trackers, and "legendary" markers leak workflow context and must never
# ship publicly. The only governance artifacts that DO ship publicly are
# CODE_OF_CONDUCT.md, SECURITY.md, CONTRIBUTING.md, and LICENSE-* — those
# are explicitly allow-listed. Everything else internal is denied here.
# Internal working dirs
.internal/
jules_tickets/
# AI/agent tool artefacts (operator-private)
CLAUDE.md
AGENTS.md
AGENT.md
agents.md
GEMINI.md
.claude/
.agents/
.codex/
.cursor/
.cursorrules
.windsurf/
.windsurfrules
.aider*
# Plans, TODOs, audits, backlogs, scoreboards, "legendary" markers
*LEGENDARY*
*_PLAN.md
*_PLAN_*.md
*_AUDIT.md
*_AUDIT_*.md
*_BACKLOG.md
*_BACKLOG_*.md
TODO.md
NOTES.md
SCRATCH.md
*_TODO.md
*_NOTES.md
ENGINE_GAPS.md
SCOREBOARD*.md
*_status_*.md
.audits/
# Allow-list: governance artifacts that ARE public
!CODE_OF_CONDUCT.md
!SECURITY.md
!CONTRIBUTING.md
!LICENSE
!LICENSE-APACHE
!LICENSE-MIT
# ── Santh-internal artifacts (END) ──