{"version":1,"name":"Agent Security Gateway","title":"Agent Security Gateway","description":"Pay-per-request security APIs for autonomous AI agents using x402. Detect Japanese prompt injection, validate content, and perform pre-payment checks.","tags":["AI","Security","Governance"],"resources":["https://agent-security-gateway.onrender.com/api/security/scan","https://agent-security-gateway.onrender.com/api/security/batch","https://agent-security-gateway.onrender.com/api/validate/deterministic","https://agent-security-gateway.onrender.com/api/security/pre-payment","https://agent-security-gateway.onrender.com/api/validate/completeness","https://agent-security-gateway.onrender.com/api/validate/list_check","https://agent-security-gateway.onrender.com/api/trust/check",{"url":"https://agent-security-gateway.onrender.com/api/tool/dry-run-validate","description":"Check tool arguments before an AI agent executes an external tool call."},{"url":"https://agent-security-gateway.onrender.com/api/tool/response-sanitize","description":"Sanitize external tool responses before they are passed back to an AI agent."},{"url":"https://agent-security-gateway.onrender.com/api/schema/drift-check","description":"Detect risky changes in MCP tool schemas, OpenAPI specs, or JSON schemas."},{"url":"https://agent-security-gateway.onrender.com/api/identity/scope-check","description":"Check whether an AI agent has the required scope for a requested action."},{"url":"https://agent-security-gateway.onrender.com/api/quota/check","description":"Check whether an AI agent is within tool call, LLM call, memory write, payment, or sub-agent limits."},{"url":"https://agent-security-gateway.onrender.com/api/security/metadata-sanitize","description":"Scan payment metadata for PII, credentials, contract details, and suspicious instructions before x402/USDC/JPYC payment transmission."}],"ownershipProofs":["0x60c402878EfcEcAe5733A88075328Aa2320C39BE"],"instructions":"Agent Security Gateway detects Japanese prompt injection and validates content. Use /api/security/scan before external API calls. Use /api/security/pre-payment before x402 payments."}