kindly-web-search-mcp-server

Pass

Audited by Gen Agent Trust Hub on May 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs users to install the 'uv' tool from its official source at 'astral.sh'. Astral is a well-known provider in the developer ecosystem.
  • [REMOTE_CODE_EXECUTION]: The server is executed using 'uvx' from the author's public GitHub repository ('github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server'). This is the intended delivery mechanism for this MCP server.
  • [COMMAND_EXECUTION]: The documentation provides standard shell commands for environment setup, dependency installation (Chromium), and launching the server.
  • [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by ingesting untrusted web content from search results and specific URLs. This risk is inherent to search tools. Evidence includes:
  • Ingestion points: Tools 'web_search' and 'get_content' fetch data from external websites.
  • Boundary markers: None explicitly defined in the tool response structure to isolate retrieved content from agent instructions.
  • Capability inventory: The skill uses headless browser automation (nodriver) to access external networks.
  • Sanitization: Employs Markdown conversion and content extraction logic to filter page elements.
Audit Metadata
Risk Level
SAFE
Analyzed
May 17, 2026, 09:50 PM
Security Audit — agent-trust-hub — kindly-web-search-mcp-server