@agenttrust-sdk/mcp 0.2.1 → 0.2.3

package/dist/embedded-docs/integration-guides/facilitator-adapters.mdx

@@ -0,0 +1,85 @@
+---
+title: Facilitator adapters
+description: Add a new x402 facilitator without rewriting AgentTrust routes or policy.
+---
+
+The adapter layer is the reason AgentTrust is not locked to Pay.sh. Pay.sh is live today; Dexter is the next worked path; atxp_ai and MCPay are marked roadmap through explicit stubs.
+
+## Mental model
+
+```txt
+HTTP request
+  -> routes/{verify,settle}
+  -> FacilitatorRegistry
+  -> active FacilitatorAdapter
+  -> VerifyContext
+  -> policy decision
+  -> settlement proof validation
+  -> feedback emission
+```
+
+Routes, policy logic, and registry reads stay unchanged. Protocol quirks live in one adapter file.
+
+## The five-method contract
+
+| Method | Why it exists |
+| --- | --- |
+| `parseRequest(req)` | translate the facilitator's request body and headers into `VerifyContext` |
+| `formatChallenge(decision, ctx)` | render Allow / Deny / RequireValidation in that facilitator's wire format |
+| `formatSettlement(ctx)` | return protocol-specific settlement metadata or an unsigned transaction skeleton |
+| `validatePaymentProof(proof, ctx)` | verify proof shape and cross-check it against the verify-time context |
+| `emitFeedback(ctx, settlement)` | emit the AgentTrust feedback record after settlement |
+
+If an adapter needs a sixth method, that is a contract change, not a one-off escape hatch.
+
+## Status map
+
+| Facilitator | Repo status | Meaning |
+| --- | --- | --- |
+| Pay.sh | concrete adapter | ready path for the demo and production dependency factory |
+| Dexter | stub / in flight | worked example for proving adapter portability |
+| atxp_ai | stub | roadmap, not silently unsupported |
+| MCPay | stub | roadmap, not silently unsupported |
+
+## Add a facilitator
+
+1. Read the facilitator's wire format. Identify challenge headers, proof headers, body shape, recipient encoding, payment ID hints, and Allow / Deny responses.
+2. Define strict Zod schemas for the request body and proof payload.
+3. Implement one `FacilitatorAdapter` class.
+4. Re-export it from `trustgate/server/src/facilitators/index.ts`.
+5. Register it with `FacilitatorRegistry`.
+6. Add unit tests that mirror Pay.sh coverage.
+7. Add a demo under `examples/<name>-demo` if the integration needs a runnable walkthrough.
+
+## Registry wiring
+
+```ts
+import {
+  FacilitatorRegistry,
+  PaySh,
+} from "@agenttrust/trustgate-server";
+
+const registry = new FacilitatorRegistry();
+registry.register(new PaySh(payShDeps));
+registry.setDefault("pay-sh");
+```
+
+Request-time selection can come from `X-Facilitator`, environment defaults, or a programmatic default. The active adapter is the only code that knows about the chosen facilitator.
+
+## Security checklist
+
+Every adapter must preserve the same proof-of-payment checks:
+
+- replay defense
+- self-pay defense
+- amount, mint, and recipient cross-checks
+- network match
+- expiry window
+- idempotent feedback emission
+- stable reason codes on Deny
+
+Pay.sh's `proof-validator.ts` and `feedback.ts` are the reference pattern.
+
+## Target integration time
+
+The architecture target is under two hours for a facilitator that already has a clear x402 wire spec. If adding a facilitator forces route edits, policy edits, or registry read changes, the adapter boundary failed and should be fixed before the integration is considered done.

package/dist/embedded-docs/integration-guides/pay-sh-adapter.mdx

@@ -0,0 +1,110 @@
+---
+title: Pay.sh adapter
+description: The live Pay.sh + AgentTrust integration path.
+---
+
+Pay.sh is the first concrete AgentTrust facilitator adapter. It translates Pay.sh's x402 headers and `PaymentRequirements` body into the `VerifyContext` that the on-chain policy path understands.
+
+Pay.sh launch reference: [Solana Foundation launch note](https://solana.com/news/solana-foundation-launches-pay-sh-in-collaboration-with-google-cloud).
+
+## What the adapter does
+
+| Stage | Pay.sh / x402 surface | AgentTrust action |
+| --- | --- | --- |
+| Initial request | no payment proof | emit `402 Payment Required` with a base64 x402 v2 envelope |
+| Challenge | `paymentRequirements.extra.agentTrust` | include payer agent, payee agent, payee recipient, policy ID, `issuedAt`, and `serviceSignature` |
+| Retry | `PAYMENT-SIGNATURE` or `X-PAYMENT` | parse proof and rebuild `VerifyContext` |
+| Policy | `VerifyContext` | run the gate decision |
+| Allow | signed payment proof | validate transfer fields, emit feedback, forward resource |
+| Deny | gate decision | return `402` with reason code and reason name |
+
+## Files to read
+
+| File | Purpose |
+| --- | --- |
+| `trustgate/server/src/facilitators/pay-sh/index.ts` | the five-method `PaySh` class |
+| `trustgate/server/src/facilitators/pay-sh/schemas.ts` | strict Zod wire schemas |
+| `trustgate/server/src/facilitators/pay-sh/sig.ts` | SERVICE challenge signature helpers |
+| `trustgate/server/src/facilitators/pay-sh/proof-validator.ts` | replay, self-pay, amount, mint, recipient checks |
+| `trustgate/server/src/facilitators/pay-sh/feedback.ts` | idempotent feedback emission |
+| `examples/pay-sh-demo/src/middleware.ts` | Express route bridge from Pay.sh retry to AgentTrust |
+
+## SERVICE-signed challenge
+
+The SERVICE emits a signed challenge when it returns the Pay.sh `402` response. The signature binds:
+
+- `issuedAt`
+- network
+- amount
+- asset
+- `payTo`
+- payer agent
+- payee agent
+- payee recipient
+- policy ID
+- payment ID hash
+
+`PaySh.parseRequest()` verifies that signature against the facilitator public key before it accepts the request. This closes the race window where a forged `paymentRequirements` envelope could be raced against the legitimate one.
+
+## Run the demo
+
+```bash
+pnpm --filter ./examples/pay-sh-demo dev
+```
+
+```bash
+pay --sandbox curl https://demo.agenttrust.tech/protected
+```
+
+The demo uses three deterministic counterparties:
+
+| Counterparty tier | Expected result |
+| --- | --- |
+| 0 | `402 Deny` with `CounterpartyTierBelowMin` |
+| 1 | `402 Deny` with `CounterpartyTierBelowMin` |
+| 3 | `200 Allow` with `X-Payment-Receipt` |
+
+Get the exact payer agent keys:
+
+```bash
+curl -s https://demo.agenttrust.tech/health | jq '.counterparties'
+```
+
+## Production wiring
+
+The demo proves the pipeline without requiring devnet RPC in CI. The server package carries the real devnet Anchor route: `trustgate/server/src/chain.ts` loads deployed program IDLs, `/verify` simulates `gate_payment`, and `/settle` delegates proof validation plus feedback emission through the active adapter.
+
+Production swaps three dependency seams:
+
+```ts
+const adapter = new PaySh({
+  signingNetwork: "solana-devnet",
+  feePayer: facilitator.publicKey,
+  validateOnChainTx,
+  emitFeedbackCpi,
+  priorEmissionLookup,
+  replayCache,
+  signDecision,
+});
+```
+
+| Dependency | Production implementation |
+| --- | --- |
+| `validateOnChainTx` | parse the confirmed SPL transfer transaction from RPC |
+| `emitFeedbackCpi` | build and send `trustgate::emit_feedback` through Anchor |
+| `priorEmissionLookup` | read the `FeedbackEmissionLog` PDA by payment hash |
+| `signDecision` | sign canonical decision bytes with the facilitator key |
+
+## Failure cases to keep
+
+Do not remove these checks when adapting Pay.sh into a production server:
+
+- reject unknown schema fields at the root
+- reject zero or u64-overflow amounts
+- reject network mismatch
+- reject `payTo` / `payeeRecipient` mismatch
+- reject expired challenges
+- reject replayed proof bindings
+- reject transfer authority equal to facilitator fee payer
+
+These checks are part of the adapter, not the route layer.

package/dist/embedded-docs/integration-guides/x402-facilitator.mdx

@@ -0,0 +1,79 @@
+---
+title: x402 facilitator
+description: Add AgentTrust decisions to an x402 facilitator without coupling to one vendor.
+---
+
+This guide is for a facilitator that already accepts x402 payment requests and wants AgentTrust decisions before settlement. If you are using Pay.sh, start with the [Pay.sh adapter guide](/integration-guides/pay-sh-adapter). If you are adding a new facilitator, start with the [adapter pattern](/integration-guides/facilitator-adapters).
+
+## Integration points
+
+| Step | Facilitator action | AgentTrust call |
+| --- | --- | --- |
+| 1 | receive x402 request or retry proof | `FacilitatorAdapter.parseRequest` |
+| 2 | translate into payer, payee, mint, amount, and policy ID | `VerifyContext` |
+| 3 | check policy | `gate_payment` or `gatePayment()` |
+| 4 | return x402 denial or validation need | `adapter.formatChallenge` |
+| 5 | settle payment | one signed transaction path |
+| 6 | emit feedback | `adapter.emitFeedback` |
+
+## Minimal Express mount
+
+```ts
+import express from "express";
+import { mountTrustGate } from "@agenttrust-sdk/trustgate/express";
+
+const app = express();
+app.use(express.json());
+
+await mountTrustGate(app, {
+  rpcUrl: process.env.SOLANA_RPC_URL!,
+  facilitatorKeypair,
+  defaultPolicyId: 1,
+  network: "solana-devnet",
+  atomicityEnforced: true,
+});
+```
+
+## Verify request shape
+
+```http
+POST /verify
+Content-Type: application/json
+
+{
+  "payerAgentAsset": "payer-agent-asset-pubkey",
+  "payeeAgentAsset": "payee-agent-asset-pubkey",
+  "amount": "1000000",
+  "mint": "mint-pubkey",
+  "policyId": 1
+}
+```
+
+For Pay.sh, the adapter reads this context from `paymentRequirements.extra.agentTrust`, verifies `serviceSignature`, derives `paymentIdHash` from `extra.memo`, and rejects mismatched `payTo` / `payeeRecipient` values.
+
+## Denial response
+
+```http
+HTTP/1.1 402 Payment Required
+X-Payment-Network: solana-devnet
+X-Agent-Trust-Decision: Deny
+X-Payment-Required: denied
+X-Payment-Reason-Code: 6
+X-Payment-Reason-Name: CounterpartyTierBelowMin
+```
+
+## Validation response
+
+```http
+HTTP/1.1 402 Payment Required
+X-Payment-Network: solana-devnet
+X-Agent-Trust-Decision: RequireValidation
+X-Payment-Required: validation
+X-Capability-Required: <32-byte-capability-hash>
+```
+
+## Settlement rule
+
+The facilitator must keep the policy gate, token transfer, and feedback emission in one signed Solana path once the settlement builder is wired. The SDK makes that requirement explicit with `atomicityEnforced: true`.
+
+Source: `trustgate/server/src/facilitators`, `trustgate/server/src/x402.ts`, `trustgate/sdk/src/express.ts`.

package/dist/embedded-docs/programs/policy-vault/counterparty-tier-policy.mdx

@@ -0,0 +1,15 @@
+---
+title: Counterparty tier policy
+description: Quantu AtomStats trust-tier and risk checks used by the PolicyVault composer.
+---
+
+`In progress`
+
+The CounterpartyTier policy reads Quantu AtomStats fields through a manual byte parser, then checks trust tier, risk score, and confidence before payment release.
+
+| Source | Path |
+| --- | --- |
+| evaluator | [`programs/policy-vault/src/policies/counterparty_tier.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/policies/counterparty_tier.rs) |
+| AtomStats parser | [`programs/policy-vault/src/ext/atom_engine.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/ext/atom_engine.rs) |
+| proof | [`programs/policy-vault/src/proofs/inv_counterparty_tier_monotone.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/proofs/inv_counterparty_tier_monotone.rs) |
+

package/dist/embedded-docs/programs/policy-vault/index.mdx

@@ -0,0 +1,68 @@
+---
+title: PolicyVault
+description: The policy composer that returns Allow, Deny, or RequireValidation.
+---
+
+PolicyVault is the decision engine for AgentTrust. It reads payer, payee, policy, velocity, kill-switch, and optional attestation data, then returns a `GateDecision`.
+
+## Composer order
+
+The order is fixed and fail-fast:
+
+| Order | Policy | Why it runs there |
+| --- | --- | --- |
+| 1 | `KillSwitch` | cheapest global stop |
+| 2 | `Spending` | pure amount and calendar bounds |
+| 3 | `Velocity` | sliding-window spend bound keyed by payer tier |
+| 4 | `CounterpartyTier` | reads payee AtomStats trust data |
+| 5 | `RequireValidation` | reads the attestation PDA last |
+
+```rust
+pub fn compose_decision(input: ComposerInput) -> ComposerResult {
+    // KillSwitch -> Spending -> Velocity -> CounterpartyTier -> RequireValidation
+    // On Allow, deltas are returned for the Anchor wrapper to apply.
+    // On Deny or RequireValidation, deltas are None.
+}
+```
+
+## Decision shape
+
+```rust
+pub enum GateDecision {
+    Allow,
+    Deny(DenyReason),
+    RequireValidation([u8; 32]),
+}
+```
+
+`DenyReason::code()` is stable for clients even though the Borsh enum order is internal to Anchor.
+
+## Policy kinds
+
+| Policy | Inputs | State mutation on Allow |
+| --- | --- | --- |
+| Spending | `amount`, UTC day, ISO week | daily and weekly counters |
+| Velocity | `amount`, slot window, payer tier | `VelocityLedger.cumulative_amount` |
+| CounterpartyTier | payee tier, risk, confidence | none |
+| RequireValidation | subject, capability hash, attestor, expiry | none |
+| KillSwitch | per-agent paused flag | none |
+
+## Safety rule
+
+The Anchor handler snapshots all accounts before composing, then applies returned deltas only when the decision is `Allow`.
+
+```rust
+match result.decision {
+    GateDecision::Allow => {
+        spending::apply_deltas(&mut ctx.accounts.policy_account, d);
+        velocity::apply_deltas(&mut ctx.accounts.velocity_ledger, d);
+    }
+    GateDecision::Deny(_) | GateDecision::RequireValidation(_) => {}
+}
+```
+
+## Formal checks
+
+<KaniProofBadge />
+
+Source: `programs/policy-vault/src/policies/composer.rs`.

package/dist/embedded-docs/programs/policy-vault/kill-switch-policy.mdx

@@ -0,0 +1,15 @@
+---
+title: KillSwitch policy
+description: Multisig-gated pause control for PolicyVault decisions.
+---
+
+`In progress`
+
+KillSwitch is the first policy in the composer. When it is paused, the composer cannot return `Allow`.
+
+| Source | Path |
+| --- | --- |
+| evaluator | [`programs/policy-vault/src/policies/killswitch.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/policies/killswitch.rs) |
+| pause instruction | [`programs/policy-vault/src/instructions/set_killswitch.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/instructions/set_killswitch.rs) |
+| proof | [`programs/policy-vault/src/proofs/inv_paused_implies_no_allow.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/proofs/inv_paused_implies_no_allow.rs) |
+

package/dist/embedded-docs/programs/policy-vault/require-validation-policy.mdx

@@ -0,0 +1,15 @@
+---
+title: RequireValidation policy
+description: Capability-attestation checks for payments that need validation evidence.
+---
+
+`In progress`
+
+RequireValidation returns `RequireValidation` when the required capability is missing, and returns `Deny` when the supplied attestation is expired, revoked, or not issued by an accepted attestor.
+
+| Source | Path |
+| --- | --- |
+| evaluator | [`programs/policy-vault/src/policies/require_validation.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/policies/require_validation.rs) |
+| attestation parser | [`programs/policy-vault/src/ext/validation_registry.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/ext/validation_registry.rs) |
+| proof | [`programs/policy-vault/src/proofs/inv_validation_expiry_correct.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/proofs/inv_validation_expiry_correct.rs) |
+

package/dist/embedded-docs/programs/policy-vault/spending-policy.mdx

@@ -0,0 +1,15 @@
+---
+title: Spending policy
+description: Per-transaction, daily, and weekly limits enforced before payment settlement.
+---
+
+`In progress`
+
+The Spending policy gates a payment against configured per-transaction, daily, and weekly limits. Allow-path deltas are applied only after the full PolicyVault composer returns `Allow`.
+
+| Source | Path |
+| --- | --- |
+| evaluator | [`programs/policy-vault/src/policies/spending.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/policies/spending.rs) |
+| composer | [`programs/policy-vault/src/policies/composer.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/policies/composer.rs) |
+| policy init | [`programs/policy-vault/src/instructions/init_policy.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/instructions/init_policy.rs) |
+

package/dist/embedded-docs/programs/policy-vault/velocity-policy.mdx

@@ -0,0 +1,15 @@
+---
+title: Velocity policy
+description: Sliding-window payment counters with tier-adjusted limits.
+---
+
+`In progress`
+
+The Velocity policy applies a rolling counter before a payment can proceed. The composer carries velocity deltas only on the all-policies-allowed branch.
+
+| Source | Path |
+| --- | --- |
+| evaluator | [`programs/policy-vault/src/policies/velocity.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/policies/velocity.rs) |
+| account state | [`programs/policy-vault/src/state/velocity_ledger.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/state/velocity_ledger.rs) |
+| proof | [`programs/policy-vault/src/proofs/inv_velocity_counter_le_limit.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/proofs/inv_velocity_counter_le_limit.rs) |
+

package/dist/embedded-docs/programs/trustgate.mdx

@@ -0,0 +1,53 @@
+---
+title: TrustGate
+description: The x402 facilitator program that emits ERC-8004 feedback through Quantu.
+---
+
+TrustGate is the on-chain feedback bridge for x402 facilitators. It owns a per-facilitator PDA, signs feedback CPI calls into Quantu, and records idempotency by payment hash.
+
+Pay.sh uses this path through the `PaySh` adapter. Other facilitators should reach the same on-chain instruction through their own adapter rather than adding facilitator-specific branches to routes.
+
+## Instructions
+
+| Instruction | Purpose |
+| --- | --- |
+| `init_authority(facilitator)` | creates the facilitator PDA |
+| `emit_feedback(...)` | CPI into Quantu `agent_registry_8004::give_feedback` |
+| `dispute_payment(...)` | emits negative-score feedback for disputed payments |
+
+## PDA signer
+
+TrustGate signs CPI with:
+
+```txt
+["trustgate_auth", facilitator_pubkey, bump]
+```
+
+That PDA is the feedback authority, not the payee agent. This prevents self-feedback while allowing a facilitator to write receipt-grade feedback after settlement.
+
+## Feedback idempotency
+
+`FeedbackEmissionLog` is initialized at:
+
+```txt
+["feedback_log", payment_id_hash]
+```
+
+Reusing the same payment hash fails account initialization, so retries cannot double-emit feedback.
+
+## x402 mapping
+
+The SDK and server map PolicyVault outcomes into response headers:
+
+| Decision | HTTP | Headers |
+| --- | --- | --- |
+| `Allow` | `200` | `X-Agent-Trust-Decision: Allow` |
+| `Deny` | `402` | reason code and reason name |
+| `RequireValidation` | `402` | required capability hash |
+
+## Source
+
+- Program: `programs/trustgate/src/lib.rs`
+- CPI wrapper: `programs/trustgate/src/ext/agent_registry.rs`
+- Server route layer: `trustgate/server/src/routes/verify.ts`
+- Pay.sh adapter: `trustgate/server/src/facilitators/pay-sh/index.ts`

package/dist/embedded-docs/programs/validation-registry.mdx

@@ -0,0 +1,49 @@
+---
+title: ValidationRegistry
+description: Capability namespaces, attestors, requests, responses, and revocations.
+---
+
+ValidationRegistry is the attestation leg of AgentTrust. PolicyVault reads `ValidationAttestation` accounts when `RequireValidation` is enabled.
+
+## Accounts
+
+| PDA | Role |
+| --- | --- |
+| `CapabilityNamespace` | names a validation capability and schema URI |
+| `AttestorProfile` | records an attestor identity URI |
+| `ValidationRequest` | opens a request for an agent and capability |
+| `ValidationAttestation` | stores a response, expiry, and revoked flag |
+
+## Instructions
+
+| Instruction | Notes |
+| --- | --- |
+| `register_namespace` | permissionless, caller computes `SHA256(name_utf8)` |
+| `register_attestor` | self-registration with display URI |
+| `request_validation` | subject owner or third party opens a request |
+| `respond_to_validation` | attestor writes the attestation |
+| `revoke_validation` | original attestor sets `revoked = true` |
+
+The program source currently exposes five instructions. The docs reserve the sixth slot for the next attestor-management extension.
+
+## Attestation message
+
+The response path uses a domain-separated message:
+
+```txt
+AGENTTRUST_ATTEST || subject || capability || payload || expires
+```
+
+PolicyVault does not trust arbitrary attestations. `RequireValidation` checks the subject, capability hash, expiry, revoked flag, and allowed attestor list.
+
+## Byte fields consumed by PolicyVault
+
+| Field | Offset |
+| --- | --- |
+| subject asset | `8` |
+| capability hash | `40` |
+| attestor | `72` |
+| expires at slot | `208` |
+| revoked flag | `216` |
+
+Source: `programs/validation-registry/src/lib.rs`.

package/dist/embedded-docs/reference/byte-offset-reference.mdx

@@ -0,0 +1,20 @@
+---
+title: Byte-offset reference
+description: Fixed offsets used by AgentTrust cross-program readers.
+---
+
+`In progress`
+
+| Account | Field | Offset |
+| --- | --- | --- |
+| `AtomStats` | `risk_score` | `549` |
+| `AtomStats` | `trust_tier` | `551` |
+| `AtomStats` | `confidence` | `557..558` |
+| `ValidationAttestation` | `subject` | `8` |
+| `ValidationAttestation` | `capability` | `40` |
+| `ValidationAttestation` | `attestor` | `72` |
+| `ValidationAttestation` | `expires` | `208` |
+| `ValidationAttestation` | `revoked` | `216` |
+
+Sources: [`programs/policy-vault/src/ext/atom_engine.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/ext/atom_engine.rs), [`programs/policy-vault/src/ext/validation_registry.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/ext/validation_registry.rs)
+

package/dist/embedded-docs/reference/changelog.mdx

@@ -0,0 +1,19 @@
+---
+title: Changelog
+description: Documentation-facing release notes for the Frontier build.
+---
+
+`In progress`
+
+| Date | Change |
+| --- | --- |
+| 2026-05-06 | Pay.sh adapter layer documented with SERVICE-signed challenges and proof validation. |
+| 2026-05-06 | `examples/pay-sh-demo` added as the Pay.sh + AgentTrust x402 walkthrough. |
+| 2026-05-06 | FacilitatorAdapter contract added for Pay.sh today and Dexter / atxp_ai / MCPay roadmap adapters. |
+| 2026-05-06 | Atomic settlement narrative updated: policy check, SPL transfer, and feedback stay tied to one settlement path. |
+| 2026-05-02 | PolicyVault composer and all five policies completed. |
+| 2026-05-02 | Five Kani invariants proved green in CI. |
+| 2026-05-02 | TrustGate Anchor program and Express x402 service shipped. |
+| 2026-05-02 | `@agenttrust-sdk/trustgate@0.1.0` published. |
+
+Source: [`execution.md`](https://github.com/agenttrust-labs/agenttrust/blob/main/execution.md)

package/dist/embedded-docs/reference/devnet-program-ids.mdx

@@ -0,0 +1,24 @@
+---
+title: Devnet program IDs
+description: Current AgentTrust devnet deployment addresses.
+---
+
+These are the pinned devnet program IDs for the Frontier build.
+
+<ProgramIdsTable />
+
+## TypeScript constants
+
+```ts
+import { DEFAULT_DEVNET_PROGRAM_IDS } from "@agenttrust-sdk/trustgate";
+```
+
+The SDK currently exposes PolicyVault and TrustGate by default. ValidationRegistry is read by PolicyVault through its account parser and is listed here for deploy verification.
+
+## Anchor declarations
+
+| Program | Source |
+| --- | --- |
+| `policy_vault` | `programs/policy-vault/src/lib.rs` |
+| `trustgate` | `programs/trustgate/src/lib.rs` |
+| `validation_registry` | `programs/validation-registry/src/lib.rs` |

package/dist/embedded-docs/reference/discriminator-constants.mdx

@@ -0,0 +1,16 @@
+---
+title: Discriminator constants
+description: Pinned instruction and account constants used across AgentTrust.
+---
+
+`In progress`
+
+| Constant | Value |
+| --- | --- |
+| Quantu `give_feedback` discriminator | `[145, 136, 123, 3, 215, 165, 98, 41]` |
+| TrustGate authority seed | `trustgate_auth` |
+| Feedback log seed | `feedback_log` |
+| Capability namespace seed | `capability` |
+| Validation attestation parser size | `290` bytes |
+
+Sources: [`programs/trustgate/src/ext/agent_registry.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/trustgate/src/ext/agent_registry.rs), [`programs/validation-registry/src/state`](https://github.com/agenttrust-labs/agenttrust/tree/main/programs/validation-registry/src/state)

package/dist/embedded-docs/reference/formal-verification.mdx

@@ -0,0 +1,19 @@
+---
+title: Formal verification
+description: Kani proof harnesses that guard PolicyVault invariants.
+---
+
+`In progress`
+
+AgentTrust runs five Kani harnesses for the current PolicyVault safety surface.
+
+| Harness | Property |
+| --- | --- |
+| `paused_implies_no_allow` | paused KillSwitch cannot allow |
+| `velocity_counter_le_limit` | allow-path counter stays within limit |
+| `counterparty_tier_monotone` | looser tier requirement cannot fail after tighter pass |
+| `validation_expiry_correct` | expired attestation cannot allow |
+| `multisig_threshold_enforced` | threshold requires distinct signing members |
+
+Sources: [`programs/policy-vault/src/proofs`](https://github.com/agenttrust-labs/agenttrust/tree/main/programs/policy-vault/src/proofs), [`.github/workflows/kani-prove.yml`](https://github.com/agenttrust-labs/agenttrust/blob/main/.github/workflows/kani-prove.yml)
+

package/dist/embedded-docs/reference/mainnet-program-ids.mdx

@@ -0,0 +1,16 @@
+---
+title: Mainnet program IDs
+description: Quantu program IDs used by AgentTrust on Solana mainnet.
+---
+
+`In progress`
+
+AgentTrust currently targets its own programs on devnet and reads Quantu trust data from the Quantu registry and Atom engine IDs below.
+
+| Program | Mainnet ID |
+| --- | --- |
+| `agent-registry-8004` | `8oo4dC4JvBLwy5tGgiH3WwK4B9PWxL9Z4XjA2jzkQMbQ` |
+| `atom-engine` | `AToMw53aiPQ8j7iHVb4fGt6nzUNxUhcPc3tbPBZuzVVb` |
+
+Source: [`docs/plan/research/02-quantu-erc8004-archaeology.md`](https://github.com/agenttrust-labs/agenttrust/blob/main/docs/plan/research/02-quantu-erc8004-archaeology.md)
+

package/dist/embedded-docs/reference/quantu-agent-registry.mdx

@@ -0,0 +1,15 @@
+---
+title: Quantu Agent Registry
+description: Quantu ERC-8004 accounts and feedback CPI used by AgentTrust.
+---
+
+`In progress`
+
+TrustGate emits feedback through Quantu `agent-registry-8004::give_feedback`, while PolicyVault reads Quantu AtomStats for counterparty policy checks.
+
+| Surface | Source |
+| --- | --- |
+| feedback CPI | [`programs/trustgate/src/ext/agent_registry.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/trustgate/src/ext/agent_registry.rs) |
+| AtomStats parser | [`programs/policy-vault/src/ext/atom_engine.rs`](https://github.com/agenttrust-labs/agenttrust/blob/main/programs/policy-vault/src/ext/atom_engine.rs) |
+| research notes | [`docs/plan/research/02-quantu-erc8004-archaeology.md`](https://github.com/agenttrust-labs/agenttrust/blob/main/docs/plan/research/02-quantu-erc8004-archaeology.md) |
+