> For the complete documentation index, see [llms.txt](https://docs.djzs.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.djzs.ai/how-a-verdict-is-made/receipts.md).

# Receipts and certificates

A free check writes nothing. A paid check writes a public, permanent record. Which record depends on how you paid.

| Paid over                 | Record                              | Where                                      | Read it back                                                |
| ------------------------- | ----------------------------------- | ------------------------------------------ | ----------------------------------------------------------- |
| HTTP (x402 or MPP)        | EAS attestation, payer as recipient | Base mainnet                               | [djzs.ai/verify](https://djzs.ai/verify), or any EAS reader |
| MCP                       | Proof-of-Logic certificate          | Irys mainnet                               | `query_pol_certificates`                                    |
| MCP, with `agent_address` | The agent's trust score             | Base mainnet, indexed by the DJZS subgraph | `query_agent_trust`                                         |

All of these are written after the verdict exists and cannot change it. A failed write is reported in the response (`eas_error`, `pol_certificate.status`, `trust_score`) and the verdict stands.

## EAS attestation (paid HTTP)

|               |                                                                                              |
| ------------- | -------------------------------------------------------------------------------------------- |
| EAS contract  | `0x4200000000000000000000000000000000000021` (Base predeploy)                                |
| Attester      | `0xfB0e11471D41f88D1eE43A1bA38d885fb6b77824`                                                 |
| Schema in use | v2, UID `0x5ef67e2b8c617635431401d2872b2a6f79eeba54a8e19fd5ecd45aceda2a2030`                 |
| Schema fields | `bytes32 intentHash, string verdict, uint8 riskScore, string[] flags, string rulesetVersion` |
| Recipient     | The paying wallet (the zero address if the settlement does not name one)                     |
| Revocable     | No. No expiry.                                                                               |

Source: `djzs-trust-mcp/src/attest-worker.ts`, `djzs-trust-mcp/src/djzs-intent.ts`. On 2026-11-20 every attestation this attester has made on Base used the v2 schema.

**Schema v3 is built, not deployed.** Under v3 a resolver contract (`DJZSVerdictResolver`, rule `DJZS-VR-v1.0`) re-derives the verdict from the attested tally and refuses an attestation it contradicts. The contracts and their tests are in the repo and run in CI, but production has no resolver address configured (`EAS_RESOLVER_ADDRESS` is unset in `djzs-trust-mcp/wrangler.toml`), so receipts attest under v2.

## The two binding keys

**`intent_hash`** is what the attestation binds. It is an EIP-712 typed-data hash anyone can recompute:

* Domain: `name` `"DJZS"`, `version` `"2"`, `chainId` 8453.
* Type `DJZSIntent`: `action`, `venue`, `market`, `side`, `thesis`, `probability_basis` (strings), `size_usd_cents`, `max_loss_usd_cents` (uint256), `exit_condition` (string), `params_hash` (bytes32).
* Mapping from your request: `toDJZSIntent` in `djzs-trust-mcp/src/djzs-intent.ts`. The route decides the `action` (`pm_trade` or `perp_trade`), never the body.

**`intent_sha256`** is `0x` + sha256 of the exact text the engine read: your intent rendered as one `key: value` line per key, keys sorted (`renderIntentText`, `djzs-trust-mcp/src/engine-adapter.ts`).

`verdict_hash` is not a binding key: it covers only the outcome ([Verdicts](https://docs.djzs.ai/how-a-verdict-is-made/verdicts.md#the-verdict-hash)).

## Checking a receipt without trusting DJZS

[djzs.ai/verify](https://djzs.ai/verify) takes the receipt JSON, reads the USDC transfer from Base, recomputes `intent_hash` from your text in the browser, and reads the attestation from EAS. It compares, in your browser, what was paid, what was asked, and what was attested.

## Proof-of-Logic certificate (paid MCP)

Every in-scope paid MCP check anchors a certificate on Irys mainnet: permanent and public. The result's `pol_certificate.status` is `anchored` (with the Irys id), `disabled` (anchoring not configured) or `error`. `query_pol_certificates` reads certificates by outcome, tag or time window ([Reference](https://docs.djzs.ai/reference/registry-tools.md)).

## Trust score (paid MCP, opt-in)

Pass `agent_address` to `verify_pm_trade` or `verify_perp_trade` and the verdict is added to that agent's record on Base, after its certificate anchors. Leave it out and nothing is written for any wallet. The record cannot be deleted.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.djzs.ai/how-a-verdict-is-made/receipts.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
