Menü
Use when reviewing an external repo or article to adopt practices into TAOM — security-vet first, map novel vs duplicative, port (never install), review, commit your changes.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
Root-cause native Bannerlord CTDs (AccessViolation in TaleWorlds.Native.dll) via Event Log offsets, offline disassembly, and live debugger forensics. No symbols needed.
Dispatch an independent Codex verification job directly via `codex exec` Bash call, then retrieve and present results
Inspect all changed files, group by logical concern, and guide atomic per-concern commits following TAOM's 50/72 rule and commit trailer convention.
Load a snapshot saved by /context-save and present it so this session can pick up where the last one left off without re-deriving decisions.
Launch parallel deep-dive agents to review completed work for quality, standards, compatibility, completeness, and cross-system data flow
Respond to a Bannerlord engine update (Steam force-bump or deliberate migration) — preserve the decompile baseline, regen, diff, re-verify bindings and creature data, control-battle.
| name | adopt-external |
| description | Use when reviewing an external repo or article to adopt practices into TAOM — security-vet first, map novel vs duplicative, port (never install), review, commit your changes. |
| allowed-tools | Read, Write, Edit, Bash, WebFetch, WebSearch |
Evaluate an external repo or article and fold the genuinely-useful parts into TAOM — without importing bloat or risk. Thin entry point; the full procedure, principles, and accumulated gotchas live in docs/ai-includes/external-repo-adoption.md — read it first.
User shares a repo/article "to see what we can adopt." One source at a time.
Nine phases — read the full procedure in the doc before starting: Identify → Security-pass-FIRST → Map novel-vs-duplicative → Tiered recommendation → Implement (port, never install) → Adversarial review → Fix + RCA → Commit MINE only → Push.
See external-repo-adoption.md § The cycle for each phase in detail. (This skill stays thin and points to the doc rather than restating it — per external-skill-ports.md "Don't amplify a rule — point to it".)
evidence-over-claims.md A.4).python tools/audit_claude_config.py --root <repo> --external fires TAOM's SkillSpector-derived categories at full severity. Heavyweight option: NVIDIA SkillSpector run static-only in an isolated venv/Docker (never installed). It supplements, never replaces, the manual read./security-scan on our own result./security-scan) — describe the category or use audit-allow:./security-scan — runs at the end of the implement phase./deep-review, /review-codex — for the adversarial-review phase on C# changes.external-skill-ports.md, evidence-over-claims.md, simplicity-criterion.md — the rules this cycle leans on.