documentation-generation

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md installation and template sections explicitly instruct fetching and executing public content (e.g., curl/irm to raw.githubusercontent.com/install-all.sh and git clone of https://github.com/j0kz/mcp-agents.wiki.git in the "Installation" and "Wiki Synchronization" sections), which pulls untrusted public repository content that the agent would ingest/execute and could therefore influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes commands that fetch and execute remote scripts at runtime — e.g., curl -fsSL https://raw.githubusercontent.com/j0kz/mcp-agents/main/install-all.sh | bash and irm https://raw.githubusercontent.com/j0kz/mcp-agents/main/install-all.ps1 | iex — and also relies on npx @j0kz/... commands which download and run package code, so these URLs/dependencies execute remote code during runtime.