mcp-recon

What is mcp-recon?

mcp-recon (formerly Tellix) is a conversational reconnaissance interface and MCP server that bridges natural language with HTTP infrastructure analysis, powered by ProjectDiscovery’s httpx and asnmap tools. It is designed for security researchers and developers who want to perform web domain reconnaissance through AI assistants like Claude Desktop.

How to use mcp-recon?

Clone the repository, build the Docker image, then add the container as an MCP server in your Claude Desktop configuration using the provided docker run command. Optionally, set a PDCP_API_KEY environment variable for ProjectDiscovery API access. Once configured, issue natural language queries such as “What’s the status code for example.com?” to trigger recon tools.

Key features of mcp-recon

• Conversational commands via natural language prompts
• MCP server integration with Claude Desktop and compatible tools
• Multiple analysis modes: lite and full recon levels
• Standalone ASN lookup for IPs, ASNs, or organizations
• Docker‑ready for isolated, secure execution
• Built‑in bug workarounds for httpx stdin/stdout leaks

Use cases of mcp-recon

• Quickly fingerprint a website’s status code, server type, and IP
• Gather comprehensive metadata including TLS certificates and headers
• Perform ASN ownership lookups from natural language questions
• Integrate reconnaissance into AI‑powered security workflows

FAQ from mcp-recon

What is the Model Context Protocol (MCP)?

MCP is an open standard developed by Anthropic that enables AI assistants to interact with external tools, systems, and data sources through a standardized interface for executing functions and retrieving resources.

What recon tools does mcp-recon provide?

It offers three tools: http_lite_recon (fast lightweight fingerprinting), http_full_recon (comprehensive metadata collection), and asn_lookup (ASN, IP, or organization queries).

How do I install and configure mcp-recon?

Clone the repository, build the Docker image with docker build -t mcp/recon -f Dockerfile ., then add the container to your MCP configuration using the provided docker run command. Optionally supply a PDCP_API_KEY for ProjectDiscovery’s API.

What are the known limitations of mcp-recon?

The underlying httpx library can leak stdin, which mcp-recon works around by shielding os.Stdin. ASN enrichment may silently fail; mcp-recon includes a fallback using the official asnmap Go library.

Does mcp-recon require external API keys?

Yes, for full functionality (especially ASN lookups) you should set a PDCP_API_KEY environment variable from ProjectDiscovery. Without it, some features may be limited.