building-mcp-servers

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of Markdown documentation and reference guides. It does not contain any executable scripts or malicious instructions.
  • [EXTERNAL_DOWNLOADS]: The guide recommends fetching official SDK documentation from the modelcontextprotocol GitHub repositories. These are well-known resources for the Model Context Protocol and are considered safe.
  • [COMMAND_EXECUTION]: Provides standard development commands for building and testing servers (e.g., npm build, dotnet run, pip install). These are intended for the developer's use during implementation.
  • [CREDENTIALS_SAFE]: Code examples use clearly labeled placeholders for sensitive information (e.g., EXAMPLE_API_KEY, your_api_key_here, GITHUB_TOKEN=ghp_xxx), adhering to secure coding practices.
  • [INDIRECT_PROMPT_INJECTION]: The skill correctly identifies the risk of indirect prompt injection in the systems it teaches users to build and explicitly provides defensive guidance, such as input validation using Zod/Pydantic and scope enforcement.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 08:29 PM
Security Audit — agent-trust-hub — building-mcp-servers