project-init

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (Step 6) instructs installing skill-manager from a public GitHub URL ("skill-manager install https://github.com/.../skills/skill-manager") and then delegates installing skills from configured skill_sources via skill-manager (bash .../scripts/install.sh "<source_path>/<skill_name>"), which clearly allows fetching and installing untrusted third‑party code/content that the agent will read/use and can alter tool behavior.