codex-cli

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly documents a --search/web-search capability (references/options.md and references/commands.md) and shows examples using --search (references/examples.md), meaning the agent can fetch and ingest open web/public content (untrusted third‑party) which would be read and used to influence analysis/decisions.