{"x402Version":2,"name":"MCP Manifest Scanner","description":"Static security scan of a Model Context Protocol server manifest: dangerous capabilities, prompt injection, hidden unicode and unconstrained parameters. Deterministic and fully offline.","version":"0.1.0","openapi":"https://mcpscan.openverbs.com/openapi.json","resources":[{"url":"https://mcpscan.openverbs.com/v1/scan","method":"POST","description":"Statically analyse a Model Context Protocol (MCP) server manifest for security risks. Tool descriptions enter the model context, a prime prompt-injection vector. Deterministic and offline (inline manifest). Flags dangerous capabilities (command/code execution, filesystem, credential and network access), injection patterns, hidden unicode and unconstrained parameters. Returns severity-ranked findings and a 0-100 score.","tags":["scan mcp manifest","mcp security audit","detect prompt injection","mcp tool risk analysis","agent tool safety check","manifest risk score"],"accepts":[{"scheme":"exact","network":"eip155:8453","price":"$0.004","payTo":"0xb3f4de103F34b99f0Da19B4F16D6B53a14C2194A","asset":"USDC"}]}]}