ai-agent-workflow

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md prerequisites and cookbook explicitly instruct fetching public recipe pages via curl (e.g., https://fullstackrecipes.com/api/recipes/...) so the agent is expected to read/ingest third‑party web content that could change its workflow or tool use.