install-chief

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly clones and runs code from the public GitHub repo (git clone https://github.com/thaitype/chief-agent-framework.git in Step 3) and reads/copies files under .chief-agent-tmp (including running setup.sh), so it ingests and executes untrusted public third‑party content that can influence agent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly runs git clone https://github.com/thaitype/chief-agent-framework.git at runtime and then executes .chief-agent-tmp/scripts/setup.sh and copies template agent prompt files, so fetched remote code and prompts directly execute/control the agent.