Every agent-payable API.

Semantic content moderation, not a word list: profanity (even disguised, any language), harassment, hate, sexual content, threats, self-harm, scams. Per-category scores, allow/review/block, configurable ignore list.

For model cascades: is the small model's answer usable, or should the larger model be paid? Detects refusal, evasion, off-topic answers, contradiction and incompleteness; decides accept, review or escalate. Does not fact-check without a source: provide the source to check substance.

Customer support triage and spam filter: classifies an inbound message by category and urgency, scores spam, and returns a routing decision (queue, escalate, review, discard).

Did a model reply leak what it should keep secret? Canary tokens and secret phrases searched by code (even split, spaced or uppercased), then paraphrases of the system prompt read by the engine. Returns clean, review or leak.

For autonomous agents: before acting, continue or stop and ask? Estimates irreversibility, ambiguity of the instruction, missing information and risk, scaled to the stakes you declare. Returns proceed, proceed_with_check or ask_human with reasons.

PII detector that finds what regex misses (names, addresses, ID numbers, health data, credentials) and locates what patterns can verify (emails, valid IBANs, Luhn-checked cards, phones). Does not rewrite text.

Before an agent installs an MCP server, skill or tool: is the description poisoned (hidden orders to the agent), does it ask for more permissions than its purpose, does it imitate a known tool? Code checks (lookalike names, broad permissions, hidden text) then a reading of the descriptions. Judges t

Reads your whole trade history and answers with exact laws, not a market forecast: is your edge proven (bootstrap, probabilistic Sharpe ratio, trades still needed), does your position size make you lose despite an edge (Kelly, volatility drag), did your edge die and when (CUSUM change detection), wh

Output guardrail for customer-support chatbots. Blocks unauthorized promises, invented facts, leaked instructions or customer data, off-topic answers and bad tone before the reply is sent. Returns send, human_review or block.

Agent security guardrail. Judges the pair (untrusted content + proposed action) before an AI agent acts: detects tool calls that obey hidden prompt injection, leave the user's goal, destroy data or send data or money elsewhere. Adjustable sensitivity and optional multi-model consensus. Returns allow

Computer-use agent helper. Send a screenshot (PNG/JPEG) and 1 to 8 element descriptions, get click coordinates (fractions and pixels) and the probability each UI element is really visible. Visual grounding for GUI automation and browser agents.

Invoice-fraud guard for AI agents that pay bills. Detects supplier bank-detail change fraud (BEC) before payment: checks the IBAN and sender domain against your supplier record (lookalike domains included), then reads pressure and off-channel requests. Returns ok, verify_callback or freeze_payment.

Inbound lead qualification and routing for sales agents: spam, job application, vendor pitch, existing customer, or a real project with urgency. Hard rules first (disposable email, tests), returns a routing lane.

Screen an EVM address on Base before sending it funds. POST the address and get back a safe/caution/danger verdict plus reasons, based on OFAC sanctions lists, known scam/blacklist databases, and on-chain age and activity. Built for autonomous agents to vet a counterparty or payout address before executing a transfer. Three tiers - quick, standard, deep - trade thoroughness for price.

Screen an EVM address on Base before sending it funds - deep tier, the service's most thorough. POST the address and get back a safe/caution/danger verdict: OFAC sanctions, known scam/blacklist databases, and deeper on-chain behaviour analysis (transfer history, activity patterns). Built for autonomous agents vetting a counterparty or payout address before a transfer. "safe" means no risk signals were found; not a guarantee the address is safe.

Verdix lite ($0.01): cheapest check of an EVM address on Base before you pay it: OFAC sanctions, scam/phishing lists, live address poisoning. Limited checks: clean = no_known_risk, never safe. Full check: /risk/address/quick.

Screen an EVM address on Base before sending it funds - quick tier. POST the address and get back a safe/caution/danger verdict: OFAC sanctions, known scam/blacklist databases, and a fast on-chain address-age check. Built for autonomous agents vetting a counterparty or payout address before a transfer. "safe" means no risk signals were found; not a guarantee the address is safe.

Screen an EVM address on Base before sending it funds - standard tier. POST the address and get back a safe/caution/danger verdict: OFAC sanctions, known scam/blacklist databases, and deeper on-chain behaviour analysis (transfer history, activity patterns). Built for autonomous agents vetting a counterparty or payout address before a transfer. "safe" means no risk signals were found; not a guarantee the address is safe.

Call when an agent needs a Verhoeff check digit from the published tables. Not an identity and not an account.

Verify one legal citation. Body: {"citation": "<string>"} - full ("Miranda v. Arizona, 384 U.S. 436 (1966)") or bare ("384 U.S. 436"); the case name enables mismatch detection. Deterministic: regex plus cache / 14.2M-case corpus / CourtListener, no LLM in the path. status=verified (name confirmed), suspicious (reporter resolves to a DIFFERENT case - the fabrication signature), or not_found. verified=false is a finding, not an error: do not retry. Only verified results are cached.

Verify up to 50 citations in one call, one flat price. Body: {"citations": ["<string>", ...]} - same shapes and verdicts as the single route. Returns results[] (one per citation), the verified/not_found/suspicious counts, verification_time_ms, hallucination_risk (LOW all verified, MEDIUM 1-2 not found, HIGH 3+). Deterministic, no LLM in the path. suspicious = a name_match mismatch: a DIFFERENT case than the name given. Do not retry a verified=false finding. Only verified results are cached.

Monitor a French company for 90 days: get alerted by signed webhook, or by polling, when a new insolvency proceeding (safeguard, receivership, liquidation) or deregistration is published in BODACC. Use it after onboarding a supplier or customer for continuous KYB and credit-risk monitoring. Input: SIREN and an optional HTTPS webhook. $0.10 per company, no account.

French company insolvency check by SIREN: current collective proceeding state (safeguard, receivership, liquidation, continuation plan, closure) and dated court events from BODACC with links to the official notices. Use it before granting credit, paying an invoice in advance or onboarding a French supplier. Cheaper than a full KYB check. Official BODACC data, errors are never charged.

Find a French company's SIREN from its name, optionally filtered by postcode or INSEE commune code. Returns up to 10 matches (name, status, NAF code, city, creation date), no personal data. First step before KYB / supplier verification of a French company.

Check RGE (Reconnu Garant de l'Environnement) energy-renovation certifications of a French establishment by SIRET: each qualification with domains, certifying body, validity dates and whether it is valid today. Source: ADEME official list. Useful to verify French contractors, artisans and installers for subsidies (MaPrimeRenov, CEE).

France: French company verification (KYB) by SIREN in one call. Legal identity, intra-EU VAT number checked live in VIES, active/closed status, age, activity, headcount, establishments, insolvency proceedings from BODACC (safeguard, receivership, liquidation, closure), RGE/Qualiopi labels. Returns an explained verdict (favorable / vigilance / unfavorable) with stable, sourced signal codes. Official open data. Supplier verification for AI agents.

Batch France company verification (KYB): up to 50 SIREN per call, same output as /v1/verify for each (status, insolvency proceedings from BODACC, explained verdict, signal codes). Usage-based: you authorize up to $0.008 x n and pay only for successful verifications. Supplier and vendor list screening for AI agents.

Global legal entity lookup in the official GLEIF LEI registry (CC0): by LEI or by national registration number (e.g. French SIREN). Use it to identify a foreign supplier, customer or counterparty and its corporate group before onboarding. Returns legal name, status, registered address, registration and renewal dates, direct and ultimate parent, and explained signals. Sole proprietors are excluded, no personal data. Errors are never charged.

Asset-freeze and sanctions screening of a legal entity against the official French national asset-freeze register (DG Tresor), which lists EU, UN and French national freezes (not OFAC or UK lists). Use it before paying, onboarding or contracting with a foreign or French company. Matches by registration number or by company name and aliases, with strong or to-verify results, the legal basis and the register publication date. Legal entities only, no personal data. Errors are never charged.

Verifiable Random pick — string selection with or without replacement bound to a future drand round the server waits for (typically 3-6s) and verifies locally, with a reproducible proof transcript.

Verifiable Random dice — unbiased dice rolls bound to a future drand round the server waits for (typically 3-6s) and verifies locally, with a reproducible proof transcript.

Verifiable Random shuffle — a uniform Fisher-Yates permutation bound to a future drand round the server waits for (typically 3-6s) and verifies locally, with original indices and a reproducible proof transcript.

Mesmo dado do /verificar-cnpj (situação cadastral, Simples Nacional, endereço) — rota técnica liquidada pelo Facilitador CDP da Coinbase, só pra indexação no Bazaar/agentic.market. Não é um produto novo.

Read a page and get a deterministic verdict on whether the r…

Verify Product data against multiple representations on the requested page. Returns field-level corroboration, bounded evidence, contradictions, observation time, and a content hash. Outcomes: verified, conflicted, or insufficient_evidence. Non-product pages return 422 and do not settle. $0.02 USDC per successful result on Base.

ABI and readable function/event signatures of a verified smart contract from Sourcify, with contract name, compiler, licence, proxy implementation and deployer. Use before calling or decoding a contract.

Verified Merchant Program eligibility.

Verified Swap Intent: two signed schema-v3 Pre-Trade gates for one same-chain swap, plus one Execution Receipt credit after x402 settlement. A BLOCK is a complete, charged decision.

Fetch the verified ABI for a contract address from Sourcify, plus a ready-to-use list of human-readable function signatures with their 4-byte selectors. Covers 8 EVM mainnets. Unverified contracts return a structured {verified:false} miss. Pair with /api/calldata-decode to decode transactions against this ABI.

Fetch the verified Solidity source files and compiler metadata for a contract address from Sourcify's open verification repository. Covers 8 EVM mainnets (ethereum, base, polygon, arbitrum, optimism, bsc, gnosis, celo). Contracts that were never verified return a structured {verified:false} miss, not an error. Feed the returned source into /api/solidity-scan for a heuristic pattern check.

Checks an EVM address against the OFAC SDN list of sanctioned digital currency addresses, refreshed daily. Use as a first compliance screen before paying or receiving. Input: adresse. Not legal advice; returns 503 if the data is older than 96 h.

Verify a block under the SuperRoot — chain 8453 block 52180190

Verify a Canadian federal corporation by name: corporation number, status (Active/Dissolved/etc.), incorporation (anniversary) date, governing legislation, and registered office address. Corporations Canada open data (OGL-Canada). Federal corporations only — excludes provincial SMEs.

Verify a Colorado business entity's status: formation standing (Good Standing / Delinquent / Dissolved), entity type, formation date, principal address, and registered agent. Colorado Secretary of State public-domain data. Colorado entities only.

Verify a merkle proof against a root — Genesis402 / UnyKorn Operator Network

Verify a single email: syntax, MX records, SMTP RCPT probe, catch-all, disposable, role-based, free-provider, typo suggestion, 0-100 score, and verdict.

Verify a single email address for deliverability. Returns a search ID to poll for results via the Fetch Search Result endpoint.

Verify a small business's SBA loan history: every SBA 7(a) and 504 loan since 1991 — amounts, approval dates, program, and outcome (paid in full, charged off, defaulted, cancelled). SBA FOIA disclosure data, federal public domain.

Verify a US business entity: status, formation date, registered agent, from official Secretary of State bulk data (plus GLEIF fallback for large companies).

Verify a US freight carrier by DOT or MC number. Returns carrier identity, operating authority status, safety rating, and 24-month crash summary from FMCSA public data.