wrong.systems

Paid x402 endpoint for multi-model AI voting and decision synthesis.

Decision procedure and verification schema for accepting a cryptographically signed AP2 payment mandate before settlement. Checks signature validity, nonce uniqueness, mandate scope against requested amount, expiry and session freshness, delegation chain depth, and parent agent authority. Returns PASS/FAIL with explicit rejection reasons for each gate.

Due-diligence audit of any x402/MPP merchant origin on behalf of a paying agent: probes /api/discovery + service catalog, does an unpaid GET on a paid service to capture its real 402 PAYMENT-REQUIRED challenge (x402 version, resource, price), checks the free meta contract, records security headers, and cross-checks registration on x402scan, PayAI discovery and ScoutScore. SSRF-guarded; only public http(s) origins.

Decision procedure for agents buying media generation pay-per-request: when to pay an image, video, or audio generation API versus free tiers or self-hosted open models. Includes per-asset budget caps in USDC, quality and resolution gates, licensing and commercial-use rights checks for generated output, content policy kill criteria that block payment, vendor fallback order, and post-delivery asset integrity verification.

Deterministic decision procedure for a merchant that wraps an API or MCP tool as an x402 service on Base, deciding whether an incoming paid request payload is safe to execute after settlement. Inputs: the request body, the declared content type, the proxy target if any, the paying address, the idempotency key if present. Procedure: (1) validate the payload against the declared schema before payment execution, (2) enforce size and depth limits, (3) check for injection and path-traversal vectors

Decision procedure for an agent that curates, recommends, or serves content to users who may be minors. Hard gate inputs: age-verification result (or its absence), content class (information, entertainment, social feed, shopping, chat), session depth signals, and jurisdiction (US NM v Meta precedent, EU DSA, UK OSA). Outputs one of three verdicts per session: refuse (minor + high-engagement-content class), throttle (session cap, recommendation throttling, dark-pattern prohibition, no infinite

Decision procedure for an agent that must choose which stablecoin to settle a payment intent in when the merchant accepts several (USDC, USDT, USDGO, and similar). Inputs: accepted denominations per merchant offer, wallet balances per denomination, facilitator and rail fee asymmetry per coin, exchange-rate volatility of non-USD-pegged or weakly-pegged assets, and the principal policy envelope on denomination preference. Outputs one verdict per intent: settle-native (pay in the coin the wallet

Agent decision procedure for x402 payments: inputs are task goal, budget remaining, latency budget, and required confidence. Auto-pay when expected value clearly exceeds price and failure is cheap to retry. Deliberate when price is material or upstream unproven. Defer when identity, refund path, or provenance missing. Keep request id, payer policy version, evidence hash, settle receipt. Handle failure modes like empty payload, wrong schema, double-charge with idempotency keys. Demands

Given a Base address, check ETH/USDC balance, transaction activity, and contract status, with a low-risk/review/caution verdict for pre-payment counterparty screening.

Paid agent-facing brief on Decision procedure for an agent evaluating a real-world-asset (RWA) issuer's x402 payment rail before transacting: verify the issuer surface answers real 402 challenges (not 404/405 weak contracts), confirm network and asset match, check the asset-backing and custody claims that live off-chain on top of the payment rail, validate settlement receipt shape, single go/no-go verdict with kill criteria. Blunt decision procedure, not marketing.

Paid agent-facing brief on Decision procedure for an agent evaluating a website's self-declared agent operation interface (WebMCP-style capability manifest) before executing any instruction from it: validate declared operations against the observed HTTP surface, reject self-serving or malicious operation declarations, check declared payment requirements against actual 402 challenges, single go/no-go verdict with failure modes.-0. 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

Given a package name and ecosystem, query OSV for known vulnerabilities and get a pinned-or-upgrade verdict an agent can act on before adopting a dependency.

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

nuclei exposure/misconfig scan of a public host for weak TLS, open services and misconfigurations.

Geocode a place and return local time, UTC offset, current weather, and an ops-suitability note - context for an agent coordinating human-dependent work across timezones.

Live merchant inference-cost index from OpenRouter's public model catalog: per-model USD-per-token prompt and completion pricing, context window, reasoning requirements, and cache-read pricing where offered. Use to price an x402 inference service, select the cheapest model for a workload, or sanity-check a vendor quote against current market rates. Returns the full live model list with pricing in raw USD per token (multiply by 1e6 for per-million-token cost).

Live counterparty due diligence for agent-to-agent commerce: given a domain, compose free public records (registrar RDAP, Certificate Transparency logs, TLS posture, DNS structure) into a trust scorecard an agent can use before paying an origin. Evidence included; priced per report.

Lightweight HTTP audit of a page: status, redirect-to-HTTPS, security headers, robots.txt, sitemap, meta - scored with actionable fixes. Useful before linking, scraping, or paying a web origin.

Paid routing endpoint: feed a payment intent (amount, optional rail filter, reliability floor) and get every eligible rail quoted with expected fee, total cost, reliability, finality, a cheapest-rail recommendation, and float sizing for pooled settlement. Machine-readable JSON computed from the provider's measured per-rail fee index.

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

Post-purchase gate for x402 web automation receipts. Checks payment match against quote, execution proof fields, result schema conformance, idempotency key reuse, and replay resistance before accepting a claimed page action as certified delivery evidence.

Weekly digest of rail economics for x402/MPP payments: per-rail transfer fees, facilitator bps, live Base gas, last-week settlement volume by rail/facilitator/product, sampled transfer gas, and an LLM-composed signal. Cache is 7 days so each paid call is a fresh weekly read.

GitHub repository health + license compliance in one call: license, stars, maintenance status, red flags (archived, stale, missing license), reuse readiness score.

Live Bitcoin mempool congestion + 's measured Base x402 transfer fee, combined into a pay-now/wait/use-Base recommendation before a microcharge settles.

Extract a webpage (title, author, text) with trafilatura and convert it into structured JSON: topic, key points, entities, summary. For agents that need clean data from a URL.

YouTube metadata intel (title, channel, duration, views, likes, description) summarized for an agent deciding whether a video is worth watching. No transcript (v1).

Scan a public domain with nuclei (OSS) for exposed vulnerabilities and get a prioritized report: findings by severity, top matches, and a verdict an agent can act on. Runs as a governed on-demand docker job (one at a time, memory-capped).