Decision Receipt API

API reference

Complete reference for the Decision Receipt API. Base URL https://decrec.summitcognitive.ai · Version 0.1.0 (spec 1.0) · HTTPS, JSON request/response bodies.

The API evaluates autonomous agent actions for admissibility, producing cryptographically signed receipts that form a tamper-evident audit chain. Each receipt captures evidence, policy evaluation, replay determinism, and an Ed25519 attestation. The canonical, always-current contract is the live OpenAPI spec at /docs (GET /v1/openapi.json).

Authentication

All write endpoints require an API key passed via the X-API-Key header. Keys are scoped to tiers that determine rate limits. Obtain one with a single call:

Request

POST /v1/signup
Content-Type: application/json

{ "email": "you@example.com" }

Response

{
  "api_key": "sk_decrec_<uuid>",
  "message": "API key issued. Include as X-API-Key header on requests."
}

If the email already has a key, the existing key is returned. New signups start on the free tier. Include the key on every authenticated request as X-API-Key: sk_decrec_<your-key> — it is also accepted as a Bearer token (Authorization: Bearer sk_decrec_<your-key>). Call GET /v1/account to see your tier, rate-limit status, and usage.

Common headers

Every response includes:

Core endpoints

Service health check. Add ?deep=true to include database connectivity (Postgres, Neo4j, Redis); returns 503 if any component is unhealthy.

{ "ok": true, "service": "summit-decision-receipt", "version": "0.1.0" }

Uptime, receipt counts (total / accepted / blocked / escalated), memory, and Node version.

The primary endpoint. Submit a claim for admissibility evaluation; returns a signed Decision Receipt. Request body is a ClaimInput:

action object

source object

Example response (abridged)

{
  "policy": {
    "verdict": "ALLOWED",
    "reason": "All policy checks passed",
    "rules": [
      { "rule": "min_sources", "passed": true, "reason": "2 sources >= 2 required" },
      { "rule": "min_source_types", "passed": true, "reason": "2 types >= 2 required" }
    ]
  },
  "replay": { "passed": true, "replay_hash": "sha256:..." },
  "receipt": {
    "receipt_id": "dr-...",
    "admissibility": { "status": "ACCEPTED" },
    "subject": { "agent": "claude-code", "repository": "acme-corp/widget-api", "pull_request": 42 },
    "evidence": { "inputs_hash": "sha256:...", "source_count": 2, "source_types": ["ci","code_review"] },
    "replay": { "deterministic": true, "replay_hash": "sha256:..." },
    "policy": { "status": "PASS", "checks": [] },
    "attestations": [
      { "signer": "decrec-server", "algorithm": "Ed25519", "signature": "base64...", "timestamp": "2026-05-30T10:35:00.000Z" }
    ],
    "chain": { "sequence": 1, "previous_receipt_hash": null }
  }
}

Verdicts: ALLOWED BLOCKED ESCALATED. Admissibility status: ACCEPTED NON_DETERMINISTIC REJECTED. See Policy, verdicts & escalation.

Submit a receipt object as the body. Returns valid (true only when all checks pass), admissibility, verification_depth: "cryptographic", and a checks[] array covering schema presence, hash integrity (recomputed against the ledger), and Ed25519 signature verification. Full walkthrough in Verifying receipts.

Evaluate a claim against a policy without persisting a receipt. Use it to test a policy before enforcing it — see Custom policies.

Returns the server's Ed25519 public key in PEM (text/plain) for offline, independent signature verification.

Totals across all evaluated receipts (accepted / blocked / escalated / non_deterministic), plus the set of agents and repositories seen.

Policy configuration

The optional policyConfig on /v1/evaluate overrides the default policy for that request:

Additional endpoints

Available beyond the core evaluation flow:

Error handling

Errors return a JSON body with an error field and, where applicable, a code and request_id.

Rate limits

Enforced per API key with a 1-hour sliding window. Current usage is reflected in the X-RateLimit-Limit and X-RateLimit-Remaining response headers.