awesome-mcp-servers-discovery

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download external resources, specifically cloning the YuzeHao2023/Awesome-MCP-Servers repository and installing various third-party management tools.
  • [REMOTE_CODE_EXECUTION]: The skill documents the use of npx to execute remote MCP server implementations directly, including both official reference servers (e.g., @modelcontextprotocol/server-filesystem) and community-built servers (e.g., @kiliczsh/mcp-mongo-server).
  • [COMMAND_EXECUTION]: Multiple shell command patterns are provided for system configuration, environment variable management, and tool interaction, such as mcp-get, mcp-cli, and yamcp.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill facilitates the agent's access to the Awesome-MCP-Servers repository, which contains over 6,000 community-contributed entries in README and Excel formats (SKILL.md).
  • Boundary markers: While the skill includes a 'Security Best Practices' section with warnings about code execution and prompt injection, the ingested repository data lacks intrinsic delimiters to prevent the agent from obeying embedded instructions.
  • Capability inventory: The skill catalogs servers with extensive capabilities including local file system access, database query execution, network fetching, and cloud platform management across all referenced scripts.
  • Sanitization: No automated sanitization or validation of the external repository's content is implemented; the skill relies on manual user review as a primary safety measure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 10:44 PM
Security Audit — agent-trust-hub — awesome-mcp-servers-discovery