codex-cli-bridge

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

• Third-party content exposure detected (high risk: 0.80). The skill intentionally enables and executes Codex CLI commands that perform web searches and can ingest external web content (see codex_executor.py's exec_with_search/_exec_command which adds the --search flag, the HOW_TO_USE.md examples showing "Enable web search", and workflow templates in agents_md_generator that include --search/--full-auto), so it will consume untrusted third‑party pages that can materially influence subsequent tool actions.