mcp-deploy

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill asks the agent to collect API keys/tokens and shows examples that embed them directly into command-line arguments and headers (e.g., Authorization Bearer, --env GITEA_TOKEN), which requires the LLM to include secret values verbatim in generated commands—an exfiltration risk.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill includes a runtime installation command that fetches and pipes remote shell code to the shell (which executes it) from https://astral.sh/uv/install.sh to install uvx, so the URL is used at runtime to retrieve and execute remote code.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill instructs installing a system-wide binary (cp to /usr/local/bin and chmod) which modifies system directories and typically requires elevated privileges, so it does push actions that can change the machine state.