cx-coding-agents

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text can enter the LLM context via Claude Code session text: the required runtime path cx logs over source ai_sessions_claude selects and returns conversation message bodies (e.g., claude_code.user_prompt / claude_code.api_response_body) which are user/assistant free-form text not authored by the operating user.