migrate-rules

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

• Third-party content exposure detected (high risk: 0.80). The skill's Guard Rails explicitly fetch and parse the public manifest at https://raw.githubusercontent.com/dev-goraebap/agent-collab/main/manifest.json and use its contents to decide and emit update/rename warnings (e.g., tell the user to run /agent-collab-updater), which means it ingests untrusted third‑party content that can influence agent behavior.