expose-agentkit-mcp

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill creates a vulnerability surface by processing untrusted external data (emails) and providing the agent with capabilities to perform actions like creating calendar events.
  • Ingestion points: Data enters the agent's context through the gmail_fetch_mails tool in the ReAct agent invocation loop within SKILL.md.
  • Boundary markers: No explicit delimiters or instructions are provided to the model to ignore potential malicious content embedded within the fetched emails.
  • Capability inventory: The skill possesses capabilities to execute network operations via MCP and modify state through the googlecalendar_create_event tool.
  • Sanitization: There is no evidence of content sanitization or validation of the email data before it is processed by the language model.
  • [EXTERNAL_DOWNLOADS]: The skill installs necessary dependencies from the official Python package registry.
  • Packages: The instructions include the installation of scalekit-sdk-python, langgraph, langchain, langchain-openai, langchain-mcp-adapters, and python-dotenv.
  • Sources: These are recognized libraries from the AI ecosystem and the vendor's official SDK.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 02:20 PM
Security Audit — agent-trust-hub — expose-agentkit-mcp