lcs-codebase-doc

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The skill runs scripts/scan.py, which reads arbitrary repository files (including outsider-authored README/docs/source text) and writes their contents—specifically TODO/FIXME/HACK matching lines—into .lcs/codebase/.codebase-scan.txt, which is then ingested by the agent for LLM context during Phase 2/4 validation.