building-agent-mcp-server
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes official and well-known Python libraries including scalekit-sdk-python, langgraph, and langchain-mcp-adapters to facilitate its core functionality.
- [CREDENTIALS_UNSAFE]: The documentation explicitly guides users to manage sensitive API keys and secrets using secure methods like environment variables and .env files, avoiding hardcoded credentials.
- [DATA_EXFILTRATION]: The skill performs network requests to the Scalekit vendor API and OpenAI services as part of its documented workflow. These operations are transparent and consistent with the skill's stated purpose of building an authenticated AI agent server.
- [PROMPT_INJECTION]: The skill's workflow involves processing external data (emails), which represents a surface for indirect prompt injection. This is documented as a standard functional risk of agents processing untrusted input rather than a malicious configuration within the skill itself.
Audit Metadata