agentpay-mcp 4.1.10 → 4.1.16

package/docs/authorized-cybersecurity-scan-profile.md

@@ -0,0 +1,56 @@
+# AgentPay MCP authorized cybersecurity-scan payment profile
+
+Paid cybersecurity MCP tools are different from normal paid APIs. A bad call can scan the wrong target, burn a budget, and create a compliance problem.
+
+AgentPay's buyer-side profile requires proof before signing:
+
+- target authorization attestation,
+- allowed-domain binding,
+- allowed scan category,
+- per-target spend cap,
+- scan-rate policy,
+- explicit human approval,
+- audit receipt language that keeps target authorization, spend cap, and x402 receipt metadata together.
+
+## Proof shape
+
+The fixture at `docs/fixtures/authorized-cybersecurity-scan-profile-agentaegis-2026-05-04.json` models an AgentAegis-style paid scan.
+
+Important fields:
+
+- `scan.target_domain`: the actual domain under test.
+- `authorization.allowed_domains`: domains the owner authorized.
+- `authorization.allowed_scan_categories`: categories permitted for that target.
+- `spend_policy.per_target_cap_usd`: hard cap for the target.
+- `spend_policy.spent_for_target_usd`: current spend for that target.
+- `rate_limit.scans_used_in_window`: anti-abuse rate state.
+- `approval_gate.approved`: human approval before x402 signing.
+- `audit_receipt.language`: receipt text retained for audit.
+
+## Failure rules
+
+The helper in `src/utils/authorized-cybersecurity-scan-profile.ts` denies payment when:
+
+- authorization is expired or not active yet,
+- target domain is missing from the attestation,
+- target domain is outside buyer policy,
+- requested scan category is not authorized,
+- requested cost exceeds buyer policy,
+- the request would exceed the per-target cap,
+- rate limit is exhausted,
+- human approval is missing,
+- audit retention is too short,
+- receipt language does not mention authorized target, spend cap, and x402 receipt.
+
+This keeps paid security tools usable without letting an agent turn x402 into an unauthorized scanner.
+
+## Verification
+
+Run:
+
+```bash
+npm run typecheck
+npm test -- authorized-cybersecurity-scan-profile
+```
+
+Expected behavior: unauthorized targets, expired attestations, cap overruns, exhausted rate limits, and missing approvals all fail closed.

package/docs/fixtures/authorized-cybersecurity-scan-profile-agentaegis-2026-05-04.json

@@ -0,0 +1,43 @@
+{
+  "schema": "agentpay-authorized-cybersecurity-scan-profile/v1",
+  "generated_at": "2026-05-04T21:25:00.000Z",
+  "scan": {
+    "tool_id": "paid.vulnerability.scan.basic",
+    "category": "vulnerability_scan",
+    "target": "https://example.com",
+    "target_domain": "example.com",
+    "requested_by_agent": "agentpay-demo-security-agent"
+  },
+  "authorization": {
+    "attestation_id": "authz-example-com-2026-05-04",
+    "granted_by": "security-owner@example.com",
+    "granted_at": "2026-05-04T21:00:00.000Z",
+    "expires_at": "2026-05-05T21:00:00.000Z",
+    "allowed_domains": ["example.com"],
+    "allowed_scan_categories": ["vulnerability_scan", "compliance_check"],
+    "proof_uri": "https://example.com/security/scan-authorization/authz-example-com-2026-05-04"
+  },
+  "spend_policy": {
+    "currency": "USD",
+    "per_target_cap_usd": 25,
+    "spent_for_target_usd": 8,
+    "requested_cost_usd": 4,
+    "x402_max_amount_required": "4000000"
+  },
+  "rate_limit": {
+    "window_seconds": 3600,
+    "max_scans_per_window": 3,
+    "scans_used_in_window": 1
+  },
+  "approval_gate": {
+    "fail_closed": true,
+    "requires_human_approval": true,
+    "approved": true,
+    "prompt": "Approve x402 payment for vulnerability scan against authorized target example.com with a 4 USD request and 25 USD per-target cap."
+  },
+  "audit_receipt": {
+    "receipt_id": "x402-scan-example-com-2026-05-04-001",
+    "retention_days": 365,
+    "language": "Paid cybersecurity scan executed for an authorized target with spend cap checks, x402 receipt metadata, rate-limit state, and target authorization retained for audit."
+  }
+}

package/docs/fixtures/chain-neutral-gateway-profile-rugmunch-2026-05-03.json

@@ -0,0 +1,43 @@
+{
+  "source": "https://x402-sol.cryptorugmuncher.workers.dev/.well-known/x402",
+  "checked_at": "2026-05-04T01:59:55Z",
+  "serviceName": "Rug Munch Intelligence",
+  "x402Version": 2,
+  "paymentHeader": "Payment-Signature",
+  "receiptHeader": "payment-response",
+  "networks": [
+    {
+      "network": "eip155:8453",
+      "name": "Base Mainnet",
+      "gateway": "https://x402-base.cryptorugmuncher.workers.dev",
+      "namespace": "eip155"
+    },
+    {
+      "network": "solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp",
+      "name": "Solana Mainnet",
+      "gateway": "https://x402-sol.cryptorugmuncher.workers.dev",
+      "namespace": "solana"
+    }
+  ],
+  "facilitator": "https://facilitator.payai.network",
+  "settlement": {
+    "custody": "facilitator",
+    "description": "Live profile names PayAI facilitator and advertises Base plus Solana gateways. Buyer agents still need settlement, receipt, and refund consistency checks before signing."
+  },
+  "trial": {
+    "enabled": false,
+    "description": "Top-level profile says no free trials while endpoint rows mention one free call. Treat this as a policy consistency check before payment."
+  },
+  "refund": {
+    "supported": true,
+    "mode": "automatic",
+    "description": "Capability list advertises auto_refund. Buyers should require receipt-level refund state before relying on it."
+  },
+  "manifests": {
+    "wellKnownX402": "https://x402-sol.cryptorugmuncher.workers.dev/.well-known/x402",
+    "glama": "https://github.com/Rug-Munch-Media-LLC/x402-gateway-solana/blob/main/glama.json",
+    "smithery": "https://github.com/Rug-Munch-Media-LLC/x402-gateway-solana/blob/main/smithery.json",
+    "mcpCatalog": "https://x402-sol.cryptorugmuncher.workers.dev/mcp",
+    "openapi": "https://x402-sol.cryptorugmuncher.workers.dev/openapi.json"
+  }
+}

package/docs/fixtures/dynamic-paid-mcp-manifest-rugmunch-2026-05-04-baseline.json

@@ -0,0 +1,67 @@
+{
+  "snapshotId": "rugmunch-x402-2026-05-04T01-38Z",
+  "sourceUrl": "https://x402-sol.cryptorugmuncher.workers.dev/.well-known/x402",
+  "capturedAt": "2026-05-04T01:45:00Z",
+  "commitSha": "c376c769fb0d3205ca9878691489e21ec87c78ff",
+  "x402Version": 2,
+  "organization": "Rug Munch Intelligence",
+  "primaryNetwork": "solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp",
+  "supportedNetworks": [
+    {
+      "network": "eip155:8453",
+      "name": "Base Mainnet",
+      "gateway": "https://x402-base.cryptorugmuncher.workers.dev"
+    },
+    {
+      "network": "solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp",
+      "name": "Solana Mainnet",
+      "gateway": "https://x402-sol.cryptorugmuncher.workers.dev"
+    }
+  ],
+  "facilitator": "https://facilitator.payai.network",
+  "hasPayTo": true,
+  "capabilities": [
+    "crypto_data",
+    "mcp_tools",
+    "multi_chain",
+    "free_trial",
+    "no_api_key",
+    "instant_settlement",
+    "auto_refund"
+  ],
+  "mcp": {
+    "totalTools": 175,
+    "totalServices": 28,
+    "protocol": "2025-03-26",
+    "catalog": "https://x402-sol.cryptorugmuncher.workers.dev/mcp"
+  },
+  "trial": {
+    "enabled": true,
+    "description": "One trial call available before paid calls"
+  },
+  "pricing": {
+    "endpointCount": 29,
+    "endpointsWithPrice": 29,
+    "endpointsWithPriceAtomic": 29,
+    "minimumPriceUsd": "$0.01",
+    "distinctPrices": [
+      "$0.01",
+      "$0.02",
+      "$0.03",
+      "$0.05"
+    ]
+  },
+  "directories": {
+    "wellKnownX402": "https://x402-sol.cryptorugmuncher.workers.dev/.well-known/x402",
+    "openapi": "https://x402-sol.cryptorugmuncher.workers.dev/openapi.json",
+    "documentation": "https://x402-sol.cryptorugmuncher.workers.dev/docs",
+    "mcpCatalog": "https://x402-sol.cryptorugmuncher.workers.dev/mcp",
+    "frameworks": {
+      "openai": "https://x402-sol.cryptorugmuncher.workers.dev/openai-tools",
+      "anthropic": "https://x402-sol.cryptorugmuncher.workers.dev/anthropic-tools",
+      "gemini": "https://x402-sol.cryptorugmuncher.workers.dev/gemini-tools",
+      "langchain": "https://x402-sol.cryptorugmuncher.workers.dev/langchain-tools",
+      "mcp": "https://x402-sol.cryptorugmuncher.workers.dev/mcp-tools"
+    }
+  }
+}

package/docs/fixtures/dynamic-paid-mcp-manifest-rugmunch-2026-05-04.json

@@ -0,0 +1,67 @@
+{
+  "snapshotId": "rugmunch-x402-2026-05-04T05-18Z",
+  "sourceUrl": "https://x402-sol.cryptorugmuncher.workers.dev/.well-known/x402",
+  "capturedAt": "2026-05-04T05:35:00Z",
+  "commitSha": "ab67483adb2585e475b5e2bcf37389a739e2ad97",
+  "x402Version": 2,
+  "organization": "Rug Munch Intelligence",
+  "primaryNetwork": "solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp",
+  "supportedNetworks": [
+    {
+      "network": "eip155:8453",
+      "name": "Base Mainnet",
+      "gateway": "https://x402-base.cryptorugmuncher.workers.dev"
+    },
+    {
+      "network": "solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp",
+      "name": "Solana Mainnet",
+      "gateway": "https://x402-sol.cryptorugmuncher.workers.dev"
+    }
+  ],
+  "facilitator": "https://facilitator.payai.network",
+  "hasPayTo": true,
+  "capabilities": [
+    "crypto_data",
+    "mcp_tools",
+    "multi_chain",
+    "free_trial",
+    "no_api_key",
+    "instant_settlement",
+    "auto_refund"
+  ],
+  "mcp": {
+    "totalTools": 184,
+    "totalServices": 28,
+    "protocol": "2025-03-26",
+    "catalog": "https://x402-sol.cryptorugmuncher.workers.dev/mcp"
+  },
+  "trial": {
+    "enabled": false,
+    "description": "No free trials - pay per call from $0.01 USDC"
+  },
+  "pricing": {
+    "endpointCount": 30,
+    "endpointsWithPrice": 30,
+    "endpointsWithPriceAtomic": 30,
+    "minimumPriceUsd": "$0.01",
+    "distinctPrices": [
+      "$0.01",
+      "$0.02",
+      "$0.03",
+      "$0.05"
+    ]
+  },
+  "directories": {
+    "wellKnownX402": "https://x402-sol.cryptorugmuncher.workers.dev/.well-known/x402",
+    "openapi": "https://x402-sol.cryptorugmuncher.workers.dev/openapi.json",
+    "documentation": "https://x402-sol.cryptorugmuncher.workers.dev/docs",
+    "mcpCatalog": "https://x402-sol.cryptorugmuncher.workers.dev/mcp",
+    "frameworks": {
+      "openai": "https://x402-sol.cryptorugmuncher.workers.dev/openai-tools",
+      "anthropic": "https://x402-sol.cryptorugmuncher.workers.dev/anthropic-tools",
+      "gemini": "https://x402-sol.cryptorugmuncher.workers.dev/gemini-tools",
+      "langchain": "https://x402-sol.cryptorugmuncher.workers.dev/langchain-tools",
+      "mcp": "https://x402-sol.cryptorugmuncher.workers.dev/mcp-tools"
+    }
+  }
+}

package/docs/fixtures/multi-ledger-receipt-xrpl-utilities-2026-05-04.json

@@ -0,0 +1,35 @@
+{
+  "schema": "agentpay-x402-multi-ledger-receipt/v1",
+  "observedAt": "2026-05-05T01:10:00.000Z",
+  "source": {
+    "name": "XRPL-Utilities MCP market signal",
+    "repo": "XRPL-Utilities/xrpl-utilities-mcp",
+    "evidenceUrl": "https://github.com/XRPL-Utilities/xrpl-utilities-mcp",
+    "pushedAt": "2026-05-05T01:02:00.000Z"
+  },
+  "payment": {
+    "x402Version": "2.x",
+    "paymentHeader": "Payment-Signature",
+    "receiptHeader": "payment-response"
+  },
+  "ledger": {
+    "label": "XRPL extension point",
+    "namespace": "xrpl",
+    "chainId": "xrpl-mainnet"
+  },
+  "settlement": {
+    "asset": "RLUSD or XRP, provider-declared",
+    "target": "unsupported-until-allowlisted"
+  },
+  "verification": {
+    "status": "unsupported_refused",
+    "verifier": "AgentPay buyer policy",
+    "checkedAt": "2026-05-05T01:10:00.000Z"
+  },
+  "boundary": {
+    "custody": "unknown",
+    "nonCustodial": false,
+    "description": "AgentPay does not sign XRPL settlement until signer, asset, target, facilitator, and receipt semantics are explicitly implemented."
+  },
+  "unsupportedLedgerRefusal": "Refused before signing: XRPL x402 rail is visible but not allowlisted. Add ledger namespace, asset parser, settlement target allowlist, verifier, receipt mapping, and non-custodial signer support before retrying."
+}

package/docs/fixtures/paid-tool-quality-threshold-strale-2026-05-04.json

@@ -0,0 +1,53 @@
+{
+  "schema": "agentpay-paid-tool-quality-proof/v1",
+  "generated_at": "2026-05-04T21:20:00.000Z",
+  "source": {
+    "name": "Strale-style scored paid MCP tool",
+    "url": "https://github.com/strale-io/strale",
+    "commit": "market-intel-2026-05-04",
+    "raw_score_field": "sqs"
+  },
+  "tool": {
+    "id": "web3.balance.lookup",
+    "name": "Web3 balance lookup",
+    "provider_id": "provider.strale.example",
+    "capability": "web3"
+  },
+  "score": {
+    "current": 92,
+    "maximum": 100,
+    "min_required": 85,
+    "measured_at": "2026-05-04T21:18:00.000Z",
+    "stale_after_seconds": 900,
+    "dimensions": {
+      "reliability": 94,
+      "availability": 91,
+      "receipt_integrity": 93,
+      "policy_fit": 89
+    }
+  },
+  "provider_health": {
+    "status": "ok",
+    "success_rate_24h": 0.98,
+    "last_success_at": "2026-05-04T21:19:00.000Z",
+    "stale_streak": 0,
+    "receipt_state": "verified"
+  },
+  "x402_payment": {
+    "scheme": "exact",
+    "network": "base",
+    "asset": "0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48",
+    "payTo": "0x1111111111111111111111111111111111111111",
+    "maxAmountRequired": "50000"
+  },
+  "approval_gate": {
+    "fail_closed": true,
+    "requires_human_approval": true,
+    "decision": "allow",
+    "reason": [
+      "Quality score is current and above buyer threshold.",
+      "Provider health snapshot is fresh and receipt verified.",
+      "x402 payment metadata matches buyer allowlist."
+    ]
+  }
+}

package/docs/fixtures/wallet-action-preflight-merx-2026-05-04.json

@@ -0,0 +1,54 @@
+{
+  "schema": "agentpay-wallet-action-preflight/v1",
+  "source": {
+    "name": "merx-mcp market signal",
+    "repo": "nicosmall503/merx-mcp",
+    "evidenceUrl": "https://github.com/nicosmall503/merx-mcp",
+    "observedAt": "2026-05-05T01:10:00.000Z"
+  },
+  "action": {
+    "kind": "resource_purchase",
+    "chainNamespace": "tvm",
+    "chainId": "tron-mainnet",
+    "asset": "TRX",
+    "amount": "12.5",
+    "recipient": "TAllowlistedRecipient111111111111111111111",
+    "nonce": "simulation-required-before-nonce-lock",
+    "irreversible": true
+  },
+  "simulation": {
+    "required": true,
+    "status": "passed",
+    "simulationId": "merx-style-tron-resource-sim-2026-05-05",
+    "expectedOutcome": "Buy bandwidth or energy for one allowlisted wallet action without transferring custody.",
+    "resourceEstimate": {
+      "feeAsset": "TRX",
+      "maxNetworkFee": "1.0",
+      "energy": 25000,
+      "bandwidth": 600
+    }
+  },
+  "policy": {
+    "perActionSpendCap": "25",
+    "dailyChainSpendCap": "100",
+    "allowedRecipients": ["TAllowlistedRecipient111111111111111111111"],
+    "allowedAssets": ["TRX", "USDT", "USDC", "USDD"],
+    "resourceCaps": {
+      "maxNetworkFee": "2.5",
+      "maxEnergy": 50000,
+      "maxBandwidth": 1000
+    },
+    "requireHumanApproval": true
+  },
+  "approvalCopy": {
+    "title": "Approve TRON wallet resource purchase?",
+    "summary": "AgentPay detected an irreversible wallet-action request. Simulation passed and policy caps are satisfied.",
+    "lineItems": [
+      "Recipient: TAllowlistedRecipient111111111111111111111",
+      "Amount: 12.5 TRX, cap 25 TRX per action",
+      "Simulation: merx-style-tron-resource-sim-2026-05-05 passed",
+      "Resource cost: max 1.0 TRX network fee, 25,000 energy, 600 bandwidth"
+    ],
+    "irreversibleWarning": "Signing will authorize an irreversible TRON resource purchase. Decline if recipient, amount, or resource estimate differs from the intended task."
+  }
+}

package/docs/paid-mcp-proxy-discovery-readiness.md

@@ -0,0 +1,70 @@
+# Paid MCP proxy and discovery readiness pack
+
+Toolstem and Cinderwright show the next buyer channel for paid tools: proxies that convert existing SaaS tools into x402-paid MCP endpoints, plus directories that compare paid services across x402, MPP, and L402.
+
+AgentPay should be easy for those surfaces to list without giving them custody, overstating ledger support, or hiding buyer policy.
+
+## Market signal
+
+Sources fetched during the May 4 market-intel cycle:
+
+- `toolstem/toolstem-proxy` describes an x402-paid proxy for Apify-backed MCP tools without Apify signup.
+- `cinderwright-ai/cinderwright-api` describes discovery data across x402, MPP, and L402, with 1,551 services and paid market-intelligence endpoints.
+
+## Paid-proxy recipe for tool authors
+
+Use this pattern when wrapping an existing API, actor, or MCP server with x402:
+
+1. Keep the upstream service token on the provider side. Do not ask buyer agents to hold pooled SaaS credentials.
+2. Publish a fresh `.well-known/x402` offer with price, asset, network, `payTo`, terms, and refund policy.
+3. Return typed 402 errors for unpaid, underpaid, wrong-network, stale-receipt, quota, and provider-health failures.
+4. Charge only after the request is valid enough to execute. Validation failures should not bill.
+5. Expose provider-health and quality fields before payment.
+6. Include idempotency keys so retries do not double charge.
+7. Make directory metadata crawlable without a private key.
+8. Let buyers bring AgentPay for approval gates, caps, allowlists, and x402 receipt logging.
+
+## AgentPay listing metadata
+
+Directories can list AgentPay with these fields:
+
+```json
+{
+  "name": "agentpay-mcp",
+  "role": "buyer-side x402 payment-control layer",
+  "package": "agentpay-mcp",
+  "mcp_name": "io.github.up2itnow0822/agentpay",
+  "settlement": "x402 with local non-custodial signing",
+  "supports": ["MCP", "x402", "human approval", "spend caps", "receipt logging"],
+  "does_not_provide": ["managed custody", "pooled SaaS token custody", "automatic non-EVM signing"],
+  "install": "npx agentpay-mcp",
+  "proofs": [
+    "docs/agentpay-five-tool-parity-proof.md",
+    "docs/agentpay-escrow-reputation-boundary.md",
+    "docs/agentpay-machine-payment-directory-listing.json",
+    "docs/x402-dynamic-paid-mcp-manifest-drift.md",
+    "docs/paid-provider-health-proof.md"
+  ]
+}
+```
+
+## Discovery insertion checklist
+
+Before submitting AgentPay to a paid-MCP or machine-payment directory, verify:
+
+- npm package name and current version.
+- GitHub repository URL.
+- MCP package identity.
+- Install command and required environment variables.
+- No private key required for `initialize` or `tools/list`.
+- x402-only wording is present.
+- Base/USDC production signing is not broadened into unsupported non-EVM claims.
+- Proof docs are linked for five-tool parity, manifest drift, provider health, quality gates, receipt normalization, and escrow boundary.
+
+## Outreach posture
+
+For Toolstem-style proxies, propose buyer-side approval and receipt guidance, not a partnership claim.
+
+For Cinderwright-style directories, propose adding AgentPay as a buyer-control tool with exact metadata and proof links.
+
+For Dexter-style five-tool flows, propose interoperability language: AgentPay can satisfy the same buyer sequence with local signing and explicit policy checks, while hosted session wallets remain a different trust model.

package/docs/paid-tool-quality-thresholds.md

@@ -0,0 +1,47 @@
+# AgentPay MCP paid-tool quality thresholds
+
+Strale's May 4 launch signal is clear: paid MCP buyers are going to ask for score gates, not only catalog listings.
+
+A paid tool directory can be useful and still be unsafe for wallet signing. The buyer needs four checks before any x402 call:
+
+1. A fresh tool-quality score with the exact field used by the directory.
+2. A provider-health snapshot for the provider that will receive payment.
+3. A minimum-quality policy owned by the buyer, not the seller.
+4. A fail-closed approval gate that blocks signing when the proof is stale, below threshold, or missing x402 metadata.
+
+## Proof shape
+
+The fixture at `docs/fixtures/paid-tool-quality-threshold-strale-2026-05-04.json` uses:
+
+- `score.current`, `score.min_required`, `score.measured_at`, and `score.stale_after_seconds`.
+- dimension scores for reliability, availability, receipt integrity, and policy fit.
+- `provider_health.status`, `success_rate_24h`, `stale_streak`, and `receipt_state`.
+- x402 payment metadata: network, asset, payTo, and max amount.
+- `approval_gate.fail_closed=true` plus a buyer approval decision.
+
+## Buyer policy
+
+AgentPay treats the buyer policy as the final authority. If a directory says `min_required=75` and the buyer policy says `minimumScore=85`, the buyer policy wins.
+
+The helper in `src/utils/paid-tool-quality-threshold.ts` denies payment when:
+
+- the score or proof is older than the buyer's maximum age,
+- the score is older than the proof's own stale window,
+- the score falls below the buyer threshold,
+- provider health is degraded or below the required success rate,
+- receipt state is not verified when verified receipts are required,
+- x402 network, asset, or payTo are outside the allowlist,
+- human approval is required but missing.
+
+That gives AgentPay parity with score-threshold catalogs while keeping signing authority on the buyer side.
+
+## Verification
+
+Run:
+
+```bash
+npm run typecheck
+npm test -- paid-tool-quality-threshold
+```
+
+Expected behavior: stale or below-threshold proofs return `decision: "deny"` and never permit x402 signing.

package/docs/post-quantum-spend-envelope-compatibility.md

@@ -0,0 +1,37 @@
+# AgentPay MCP post-quantum spend-envelope compatibility assessment
+
+PQSafe's May 4 signal pushes spend authorization into a post-quantum envelope frame: ML-DSA-65 signatures, AP2 and ACP compatibility language, MCP endpoints, audit ledgers, and x402 routing.
+
+AgentPay should answer that without overclaiming. Current AgentPay artifacts map cleanly to envelope concepts at the policy and receipt layer, but this package does not claim post-quantum cryptography.
+
+## Compatibility map
+
+| AgentPay control | Envelope concept | Status |
+| --- | --- | --- |
+| Spend limits | maximum authorized spend inside a signed envelope | Compatible |
+| Network, asset, and recipient allowlists | route constraints attached to spend intent | Compatible |
+| x402 receipts | receipt pointer for audit and reconciliation | Requires adapter |
+| Approval gates | approval before spend execution | Compatible |
+| agent_id, task_id, policy_version, receipt_id | audit ledger payload | Requires adapter |
+
+## Explicit non-claims
+
+AgentPay MCP does not claim any of the following from this assessment:
+
+- ML-DSA-65 signing,
+- post-quantum key lifecycle,
+- AP2 envelope conformance,
+- ACP envelope conformance,
+- Arbitrum audit-ledger publication.
+
+Those claims require implementation, fixtures, and cryptographic tests before they belong in product docs.
+
+## Practical buyer response
+
+If a buyer asks whether AgentPay can coexist with post-quantum spend envelopes, the current answer is:
+
+- AgentPay spend policies, allowlists, approval gates, x402 receipts, and audit metadata are compatible inputs to an envelope adapter.
+- AgentPay can preserve receipt IDs and policy versions so an envelope signer can bind payment approval to settlement evidence.
+- AgentPay should not market ML-DSA or post-quantum signing until tests verify it.
+
+The helper at `src/utils/post-quantum-spend-envelope-compatibility.ts` returns this assessment in code so docs and future adapter work start from the same boundaries.