codex

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly enables Codex's web search (e.g., the SKILL.md lines "All codex commands use --enable web_search_cached" and the codex exec/review commands in Step 2A/2B/2C) so the agent will fetch and consume public web/docs content during reviews, and that external content is read and used to produce verdicts and follow-up actions (e.g., PASS/FAIL gates), creating a clear indirect prompt-injection exposure.