Send branch evidence
Progress signals, verified artifacts, cost, tool calls, risk events, authorization budget, and the pinned policy reference.
A deterministic decision gate for autonomous agents. Evaluate progress, loops, budget, and evidence—then receive a portable, tamper-evident execution receipt.
Internal self-checks share the same context and incentives as the branch they judge. Agent Decision Gate supplies an independent policy boundary and a receipt another runtime can verify.
Progress signals, verified artifacts, cost, tool calls, risk events, authorization budget, and the pinned policy reference.
Continue, prune, stop, abstain, or escalate—plus reason codes and remediation that a calling agent can execute directly.
Persist the canonical request, decision, policy, and receipt with a public integrity view and branch-chain verification.
Start with one evaluation endpoint. Discovery, pricing, examples, and verification are free machine-readable surfaces.
/v1/branches/evaluateEvaluate a branch and return an action, reason codes, remediation, and receipt. Add ?persist=true for D1/R2 persistence.
/v1/receipts/verifyRecompute receipt integrity and, when stored, verify the D1 registration, R2 payload, and previous chain link.
/.well-known/agent.jsonCanonical machine entry point with absolute links, capabilities, trust boundaries, and integration routes.
/v1/openapi.jsonOpenAPI 3.1 contract for tool generation, autonomous integration, request validation, and endpoint discovery.
The service makes narrow, testable claims. That boundary is part of the product—not a disclaimer hidden after the fact.
Agents can inspect the service, generate a client from OpenAPI, evaluate a branch, and verify the returned receipt using public routes.
$ curl https://agentdecisiongate.com/.well-known/agent.json $ curl https://agentdecisiongate.com/v1/openapi.json $ curl https://agentdecisiongate.com/v1/pricing $ curl https://agentdecisiongate.com/health