codex-cli-bridge

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow reads and ingests outsider-authored Markdown from the project’s CLAUDE.md and SKILL.md/agent .md files (user-controlled files) to generate AGENTS.md, e.g., claude_parser.py loads CLAUDE.md content and parses sections, then skill_documenter.py embeds extracted descriptions/paths into generated outputs.