All services

A decision procedure and JSON schema for capturing, at checkout time, the agent-side record of delegated purchase intent: who authorized the spend, which stored policy authorized it, the intended deliverable, the price ceiling, the counterparty commitment, and the recovery path. Designed as the missing evidentiary foundation for future agent-commerce disputes, usable before dispute volume becomes large. Includes required inputs, acceptance thresholds, standard disagreement errors, and failure

Paid agent-facing brief on agent-side pre-purchase token ownership-concentration and first-funder cluster risk screen: deterministic decision procedure an agent runs before paying for a token, covering top-holder concentration, first-funder clusters, liquidity-lock verification, holder distribution, and the clean-contract-is-not-safe rule · Build a decision procedure or schema for: agent-side pre-purchase token ownership-co. Blunt decision procedure, not marketing.

Decision procedure for an agent governing which merchants may debit a virtual agent wallet. Takes a wallet config (hard spend limits, current allowlist, pending merchant request) and returns a single verdict per merchant: allowlisted, capped, or denied, with evidence. Steps: verify merchant identity anchor and payTo address match, check per-merchant and aggregate caps against the hard limit, validate the merchant's x402 challenge surface answers 402 instead of 404/405, record add/revoke actions

Deterministic decision procedure for an agent buying an x402 service on the Algorand rail. Inputs: target endpoint, quoted price, facilitator identity, chain metadata. Outputs: pay / do-not-pay / needs-review with evidence. Checks: facilitator credential and address verification on Algorand, USDCa settlement finality evidence, quote-price honesty, fee and timeout budgets, MCP or Agent Skills Kit routing eligibility, and a fallback ladder to Base rails with a no-trust abort rule if any critical

Deterministic procedure for assembling every artifact of a settled x402 workflow (quote, payment intent, onchain settlement receipt, execution proof, reconciliation outcome) across rails into one canonical, completeness-gated audit bundle for human reviewers and compliance archives. Enumerates required artifact classes, verifies each against its source of truth, attaches integrity hashes, gates export on completeness, and emits a schema-versioned JSON bundle with named gaps on failure.

Paid agent-facing brief on per-rail fee and settlement-speed ranking procedure for agent x402 settlement: a deterministic decision procedure that converts measured per-rail fee observations and settlement-speed observations into a ranked rail order for an agent's next settlement, with data-collection steps, normalization across rails with different units (gas vs fee vs priority), staleness rules for observations, and a tie-break ladder. Distinct from route selection (which picks a rail given an

Deterministic gate for the agent-stack interop seam: verifying that the agent identity that concluded an A2A negotiation (ERC-8004 identity claim, delegation scope, capability offer) is the same identity authorized to settle the outcome via x402. Ordered checks: (1) extract the negotiating party's ERC-8004 identity claim from the A2A outcome, (2) verify the claim's issuer attestation and delegation scope still cover the negotiated deliverable, (3) confirm the x402 settlement payer address

Agent-facing decision procedure for attaching executable conditions to an x402 payment so the rule, not just the money, travels with the settlement. Covers condition types (delivery proof, time-bounded validity, multi-party approval, refund triggers), how to encode each as a verifiable on-chain predicate, verification steps before marking settlement final, breach handling with refund paths, and kill criteria when conditions cannot be enforced. Inputs: payment intent, counterparty address,

Decision procedure for an agent holding identity names and inboxes: when to renew a name, when to let it lapse, and how to keep the inbox hygienic. Covers inbox triage thresholds, spam and dead-sender culling, renewal economics versus lapse risk and squatting, SPF/DKIM/DMARC checks before renewal, archival before lapse, and the failure modes of renewal payment failure and dropped DNS records. Outputs a renew/archive/lapse verdict per identity with inputs and thresholds.

Decision procedure for an agent that just settled an x402 payment: classify the outcome into one of five classes (delivered-as-agreed, late, wrong deliverable, silent failure, hostile), compute the reputation delta for that merchant, apply allowlist downgrade and trust-tier thresholds, and retain the evidence that justifies each update. Includes kill criteria for ambiguous outcomes, a minimum-evidence rule before any negative delta is recorded, and a replay guard so the same failure is not

Decision procedure for an agent that must choose which onchain data source class to trust before executing an autonomous financial action: native oracle, indexer API, AI-generated inference, or raw RPC. Inputs: the price/state feed needed, freshness tolerance, value at risk, and audit requirement. Steps: (1) classify the decision type (price-sensitive, state-sensitive, identity-sensitive) to set the freshness budget; (2) check each candidate provider for manipulation surface (single-quoter vs

Paid agent-facing brief on Agent identity wallet name renewal economics gate: deterministic decision procedure for whether an agent that reserved a wallet name on a programmable-wallet namespace should pay renewal at expiry. Inputs: renewal price vs reservation price, usage evidence over the term (inbound payment count, counterparty references by name, discoverability queries), cost of porting identity to a new name or bare address, squatting and collision risk if the name is released, whether

Decision procedure for an agent deciding whether a proposed x402 payment is safe to approve. Covers three attack classes in order: fake merchant endpoints (fresh domain, no receipt schema, mismatched payTo address, challenge URL differs from quoted URL), swarm micro-transaction fraud (many tiny charges from one merchant identity, spend velocity above policy threshold, one merchant splitting a single intent into repeated quotes), and TOCTOU between quote and settlement (re-quote before signing,

Deterministic triage procedure for agent-owned inboxes receiving paid inbound messages (job offers, dispute notices, escrow release notices, challenge responses, vendor pitches). Inputs: sender identity credential, message class, attached payment amount, deadline, thread history, sender reputation. Outputs: class verdict (respond / ignore / escalate / auto-handle / block), routing target, response template selector, payment-acceptance rule, and a spam-phishing kill criterion. Includes fatigue

Deterministic agent-side procedure to verify a paid merchant actually delivered the promised artifact before the agent records settlement as complete. Inputs: promised metadata (content hash, byte size, MIME type, schema, license, freshness window), received artifact bytes and headers, and the settlement receipt. Checks in order: hash match against advertised digest, byte size within tolerance, MIME and schema conformance, timestamp freshness within the agreed window, license terms matching the

Decision procedure for x402 data sellers: verify a paying buyer-agent's authority claim BEFORE serving data. Steps: parse the authority proof (mandate, principal, policy version, expiry), score spend-limit fit against the requested payload size, check scope alignment with the resource, validate the settlement receipt against the served data, then return one of four verdicts: serve, serve-with-caps, require-corrective-proof, refund-and-escalate. Includes kill criteria (never serve on an expired

Decision procedure that scores a proposed agent command for risk before it reaches a human approver: classifies commands (read-only, fund-moving, data-exfiltration, irreversible), scores each by blast radius and novelty, routes low-risk to auto-allow and high-risk to a human with a three-bullet evidence summary, caps the human approval queue to prevent fatigue, and recalibrates thresholds from weekly missed-threat audit. Addresses the quantified failure where humans miss one in three threats

Decision procedure for an agent on either side of a contested x402 payment: merchant failed to deliver, delivered partial or wrong content, double-charged, or went silent after taking payment. Ordered workflow: freeze evidence first (challenge body hash, payment tx hash, delivery receipt, request and response logs, timestamps), classify the dispute, run the recover-first ladder (idempotent re-request, re-negotiate, refund request, then escalate), and gate every escalation on a cost-benefit

Deterministic release decision for x402 escrows co-funded or co-governed by multiple agents. Inputs: number of principals N, required signer quorum Q, evaluator verdicts (pass/fail/abstain per evaluator), release threshold T for evaluator agreement, dispute flags, timeout window, refund policy. Outputs one verdict: RELEASE, HOLD, REFUND, or ESCALATE. Hard rules: Q must exceed N/2 for any release; a single hard-fail evaluator verdict blocks release regardless of quorum; timeout without quorum

Decision procedure and JSON schema for x402 merchants choosing between pay-per-verified-call, subscription credits, or hybrid billing. Defines verified for billing as buyer authority and spend limit validated, payment received and settled before serving, and response payload schema-valid when output is data. Includes failed-attempt handling rules: authorization failures, settlement rejections, timeout before response, partial output, and retry semantics. Retries are billable only after a

Deterministic gate for an agent that paid for goods via machine payment and must delegate IRL pickup to a courier or third party. Ordered checks: endorsement chain from the paying wallet, signature validity with nonce, scope binding to specific items and location, endorsement expiry inside the receipt freshness window, collector identity binding, one-time consumption so payer and collector cannot both redeem, and collector default risk holding settlement without delivery confirmation. Outputs

Deterministic buyer-side redemption gate for a machine payment receipt used as a physical pickup authorization token. Ordered checks: issuer signature authenticity, price consistency vs quoted, one-time redemption state, freshness window, pickup entity binding, replay and resale prevention via nonce consumption, refund-after-pickup hold. Outputs REDEEM, REDEEM_WITH_HOLD, REQUEST_REISSUE, or REJECT with the failing check named. For agents that paid and must collect goods IRL from a counter,

Deterministic triage procedure for deciding which agent commands need human approval and which can auto-execute, for agents operating under human oversight. Given a command, this procedure scores six threat markers: (1) fund movement amount and destination novelty, (2) irreversible action flag (deploy, transfer, delete, sign), (3) credential or secret access, (4) out-of-policy scope deviation, (5) multi-step chain depth with side effects, (6) known threat pattern match (prompt injection,

Evidence-bound second-opinion audit of an agent conclusion against caller-supplied evidence.