mcp-graveyard

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill reads local session logs and config files and instructs the agent to "print output verbatim," which will expose any API keys, tokens, cookies, or other secrets present in those files directly in the agent's output.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs running "npx mcp-graveyard@latest", which at runtime fetches and executes the mcp-graveyard package from the npm registry (e.g. https://registry.npmjs.org/mcp-graveyard), so it relies on and executes remote code fetched during runtime.