litellm-agent-platform

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow for this skill is to run an agent with user-provided input/system_prompt and optionally use registered MCP tools (e.g., GitHub/custom MCP); those tools can return arbitrary free-form text from outsider-authored sources (e.g., PR/issue comments or scraped content) that the agent then ingests into its LLM context via the MCP tool responses.

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

• Potentially malicious external URL detected (high risk: 0.90). The installation instructions explicitly run git clone https://github.com/LiteLLM-Labs/litellm-agent-platform.git followed by npm install and npm run (build/start), which fetches remote code and executes it as a required dependency.