darknet-mcp-server-threat-intelligence

Warn

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The installation instructions direct users to fetch code from an external GitHub repository (badchars/darknet-mcp-server) and an NPM package (darknet-mcp-server). These sources are not identified as trusted vendors or well-known services.
  • [REMOTE_CODE_EXECUTION]: The skill's primary functionality involves running an MCP server via npx or local script execution (bun run src/index.ts), which results in the execution of third-party code on the host system.
  • [COMMAND_EXECUTION]: Setup instructions include the use of sudo apt install tor and other shell commands to manage system services and install dependencies.
  • [PROMPT_INJECTION]: The skill provides tools designed to fetch and scrape content from .onion sites and dark web forums. This data is untrusted and could contain hidden instructions aimed at influencing the agent's behavior (Indirect Prompt Injection). The skill lacks explicit boundary markers or instructions for the agent to sanitize or ignore embedded commands within this external data.
  • [DATA_EXFILTRATION]: The skill requires users to provide numerous sensitive API keys (e.g., HIBP, IntelX, Hudson Rock) as environment variables. While necessary for the tool's function, this creates a high-value target for exfiltration if the underlying server code were compromised.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 1, 2026, 06:14 PM
Security Audit — agent-trust-hub — darknet-mcp-server-threat-intelligence