ln-646-project-structure-auditor

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly reads and interprets arbitrary project files from the provided codebase_root (e.g., Phase 1 “Read docs/project/tech_stack.md”, package.json / pyproject / go.mod, .env and all source files via multiple Glob checks) and uses that content to detect the tech stack and drive audit decisions, so untrusted repository content can materially influence its behavior.