{"uid":"cap_wgrrk25ahleODvo5mqR8P","slug":"goldkey-guard-execution-path-authorization-8af30c92","name":"GoldKey Guard Execution-Path Authorization","description":"GoldKey Guard is an execution-path authorization layer for AI agents: operator-signed policy, a customer-controlled local enforcer, and signed receipts before MCP, HTTPS, or supported EVM calls run.","url":"https://goldkey-edge-storefront.noah-ing.workers.dev/v1/paygo/execute","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"tool":{"enum":["json.canonicalize","json.validate","security.prompt_scan","security.url_check","policy.spend_check","text.normalize","action.gate"],"type":"string"},"input":{"type":"object"}}},"responseSchema":{"type":"json","example":{"tool":"security.prompt_scan","result":{"risk_score":50},"tool_version":"1.0.0"}},"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.01","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.01/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.01","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.01","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_foezcS6KIKotMPTPV5SpG","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.01","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Pre-authorizes an AI agent tool call (MCP, HTTPS, or EVM) against a signed operator policy and returns a signed receipt required before the call can execute","exampleAgentPrompt":"Before you run that MCP tool call or payment, check it against GoldKey Guard — scan the 'security.prompt_scan' tool call and get back a signed receipt showing whether it's ALLOW, REVIEW, or BLOCK before actually executing anything.","exampleUseCases":[{"title":"Gating a risky agent payment action","prompt":"Before my agent goes ahead and executes that USDC transfer on-chain, run a GoldKey Guard preflight check on the payment tool call and make sure we get back a signed ALLOW receipt — don't let it proceed if it's BLOCK or REVIEW."},{"title":"Enforcing operator policy on MCP tool calls","prompt":"My operator has a signed GoldKey policy that restricts which MCP tools my agent can call. Before the agent calls 'files.write', submit it to GoldKey Guard and tell me the risk score and whether it came back ALLOW or BLOCK."},{"title":"Auditing agent actions with signed receipts","prompt":"Every time my agent tries to call an external API or tool, I want GoldKey Guard to authorize it first and return a signed receipt I can log — can you run the preflight check on the 'web.fetch' tool call right now and give me the receipt and decision?"}],"resultDescription":"A JSON response containing the tool name, tool version, and a result object with a risk_score (0–100) and an authorization decision (ALLOW, REVIEW, or BLOCK), along with a signed receipt that must be presented before the guarded call is permitted to execute.","failureModes":["Missing or invalid operator-signed policy returns an authorization error","Tool call parameters that don't match the policy schema return a BLOCK decision","Network timeout to the edge worker returns no receipt, halting execution","Malformed request body returns a 4xx error with no receipt issued","Expired or revoked policy signature causes rejection","REVIEW decision requires human-in-the-loop resolution before execution can proceed"],"whenToPreferThis":"Choose GoldKey Guard when you need a cryptographically accountable preflight gate — with a signed receipt as proof — before an AI agent executes any MCP tool call, HTTPS action, or EVM transaction. It is especially suited for operator-controlled multi-agent environments where policy enforcement, audit trails, and human override (REVIEW) are required. Prefer this over simple API key auth when you need a verifiable, per-call authorization record rather than session-level authentication.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-16T00:35:31.417Z","isFirstParty":false}