kindly-web-search-mcp-server

Fail

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The installation instructions fetch and execute a setup script for the 'uv' tool from 'astral.sh'. Astral is a well-known and established service provider in the developer tools ecosystem.
  • [REMOTE_CODE_EXECUTION]: The setup instructions direct the agent to use 'uvx' to download and run code directly from an unverified third-party GitHub repository ('github.com/Shelpuk-AI-Technology-Consulting/kindly-web-search-mcp-server'). This facilitates the execution of remote code from a source that is not recognized as a trusted organization or well-known service.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes tools designed to fetch and process data from external, untrusted sources on the web.
  • Ingestion points: The web_search and get_content tools ingest content from arbitrary websites, including StackOverflow, GitHub, and Wikipedia.
  • Boundary markers: The tool definitions do not provide explicit boundary markers or instructions to the agent to ignore embedded commands within the fetched content.
  • Capability inventory: The skill returns the extracted page content directly to the agent's context.
  • Sanitization: While the skill documentation mentions intelligent extraction and conversion to Markdown, it does not detail specific sanitization logic to prevent adversarial instructions in the source data from influencing the agent.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the Chromium browser as a dependency for its web-rendering and content-extraction capabilities.
Recommendations
  • HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 16, 2026, 10:43 PM
Security Audit — agent-trust-hub — kindly-web-search-mcp-server