Every agent-payable API.

US safety recalls for a vehicle make, model and year from NHTSA: campaign number, date, component, summary, consequence, remedy and do-not-drive or park-outside warnings. Use before buying or servicing.

NHTSA safety recalls for a vehicle by make/model/year: campaign number, affected component, summary, consequence, and remedy. Live gov data, no key. ?make=honda&model=accord&year=2019

Embeds one query and up to 64 candidates locally, then returns cosine scores and stable descending ranks.

Embeds one request and a provided tool list locally, then returns deterministic similarity ranks and the selected tool.

Calculates item co-occurrence and item similarity from the submitted events only. No user profile is stored.

Embeds up to 32 texts on the local Ollama runtime. Returns explicit model and dimensions; no paid third-party API is called.

Embeds up to 32 texts on the local Ollama runtime. Returns explicit model and dimensions; no paid third-party API is called.

Screen a vendor for OFAC sanctions and complete vendor due diligence and company verification. This citation-first Vendor Risk Brief checks exact normalized names against current U.S. Treasury OFAC SDN and Consolidated primary-name and alias files, then adds GLEIF legal-entity records, SEC EDGAR company filings, and authoritative RDAP domain-registration evidence. It returns match status, vendor risk signals, identity gaps, domain-age signals, source URLs, evidence timestamps, confidence, and limitations for vendor onboarding, supplier risk, procurement, AML, and counterparty preflight. A name match requires human review; a no-match result is not sanctions clearance or a compliance determination.

Structural IBAN (MOD-97) and LEI-format checks plus a network-wide settle count. Not an identity, credit or fraud assessment; no per-counterparty history is held.

Is this vendor or domain safe? One call runs cert, DNS, email security, ransomware leak-site listing and vendor status checks and returns a verdict (safe, caution, risk, incomplete). $0.04 USDC on Base. Not charged if every check fails or input is invalid.

Extract structured text, summaries, or metadata from supplied document text or a public HTTP(S) document URL.

Compare two supplied public-source text versions and return bounded, hash-backed evidence of change.

Investor due-diligence prep — data-room structure, what VCs verify at each stage, deal-killing red flags and timeline expectations. For founder and startup agents.

Cap-table dilution modeler — post-round ownership, dilution percentages, exit-scenario modeling and benchmark comparisons. For founder and startup agents.

Term-sheet decoder — plain-English explanation of every clause, founder-hostile red flags, market-norm comparisons and negotiation guidance. For founder agents.

Startup legal-formation guide — country-specific structure, formation steps, costs, key documents and investor tax incentives (SEIS/EIS/QSBS) across 15+ jurisdictions. For founder agents.

TAM/SAM/SOM market sizing that withstands VC scrutiny — bottom-up and top-down, comparable data, stress-test questions and how to defend your numbers. For founder agents.

Pitch-deck scorer against VC frameworks (Sequoia, a16z, YC) across 8 dimensions — scores, improvement notes and the exact questions VCs will ask. For founder agents.

The broker view of a venue (a contract or wallet users pay into): every wallet that transacted with it in the window, the value segments of the book, VIP candidates by volume, which wallets went quiet (last transfer before the final third of the window), and which funding wallets carry a curated flag. Built for the 'user & VIP intelligence', 'transaction monitoring / KYC-KYB' and 'who is behind ...

Hyperliquid execution quality measured on our own orders: slippage vs mid (p50/p90 bps), taker fee actually paid, wall-clock to fill, fill rate — for perps and xyz tokenized equities — plus a liquidity snapshot (spread, round-trip cost, depth within 1%, funding APR) for the 25 deepest assets. Thousands of real IOC fills, refreshed continuously.

Vercel AI Gateway pricing table: per-token cost of every AI model Vercel AI Gateway serves, in one call — input, output, cache and batch USD per 1M tokens, ranked cheapest first, with context window and capability flags joined in. Compare LLM token cost inside Vercel AI Gateway and against other providers hosting the same model. Normalized from public sources, cross-checked, refreshed hourly.

Before an agent opens or follows a link: is this address a trap? Code finds the technical signs (IP address, deceptive characters, hidden at-sign, shortened link, lookalike of the expected brand), then the engine reads what code cannot see: imitation of a known brand and a message that pushes to cli

Before handing a user message to an assistant: does it try to push the assistant off its task or to pull out its hidden setup text? Code spots known phrases, encoded blocks and invisible characters; the engine reads disguised requests (role-play, fiction, rewordings). Returns allow, review or block.

Before an agent trusts a sender: is this email address suspicious? Code compares the domain with the expected one (lookalike, other domain), spots disposable addresses, deceptive characters and free mailboxes used in the name of an organisation; the engine judges whether the display name and subject

One-call verdict on any ERC20 on Base, Ethereum, Polygon, Arbitrum or Optimism: a machine-readable label for who can act on your balance (no-owner-powers, owner-can-intervene, owner-controls-balances, not-a-token), a headline, a plain-language reading and a paste-ready card - with the deterministic evidence underneath, selector by selector. Label, score and card come from the deployed bytecode and live state, never from a model. Research, not investment advice.

Is this token any good? Grade BASED/MID/COOKED/RADIOACTIVE, 0-100 score, one-liner, roast and reasons from live data: honeypot buy/sell simulation and taxes, owner powers (Base) or mint/freeze authority (Solana), liquidity, volume, buys/sells and pool age across all DEX pools. Body {"tokenAddress":"0x..."} or {"chain":"solana","tokenAddress":"<mint>"}. Synchronous.

Verdict API: independent verdict (clear / caution / block / unknown) on an address, contract, agent, domain or transaction — Genesis402 / UnyKorn Operator Network

Rule-based long/short/watch/avoid verdict for one trade.xyz stock perp, combining perp-vs-stock gap, basis, funding and SEC EDGAR insider/8-K/13D/dilution filings (14d). Data, not financial advice.

An independent yes-or-no style verdict, with a short reason, on a men's clothing item or outfit. Called by an agent before purchase. No affiliate cut, no placement fees.

One-line safety verdict (GO / CAUTION / HONEYPOT_RISK) for (owner/name) — impulse-priced pre-flight check

For coding agents that run shell commands: approve, review or block. Text rules first (root deletion, download-and-run, key exfiltration, reverse shells, SQL destruction, persistence, privilege escalation), then a fake terminal that simulates the command without ever running it (decoy files, pipe an

Tool results (MCP servers, APIs, search, file reads) are untrusted content the agent re-reads. Detects a result that gives orders to the agent, strays from what the tool should return or asks to send data (tool poisoning). Returns allow, review or block.

Before an agent clicks "I accept": flags risky clauses in terms of service or contracts (auto-renewal, data resale, waiver of court or class action, liability cap, unilateral changes, exit penalties, IP grab). Flags, does not give legal advice. Returns accept, review_clauses or do_not_accept.

Hallucination check for RAG: is a claim supported by the source text? Returns supported, unsupported, contradicted or uncertain with probabilities, so an agent can verify an answer before showing it.

Sentiment from -1 to +1 with confidence, mixed-opinion flag and a separate score per aspect you name (price, support, delivery). Abstains when the text is ambiguous.

Pre-trade risk check for traders and trading bots. Before an order is sent, checks it exactly against your own rules (risk per trade, daily loss, max loss, cooldown after a loss, trade quota, blocked hours, days and instruments, minimum reward to risk) and computes the largest size still allowed. If

Pre-flight guardrail for untrusted text before an LLM agent reads it: prompt injection, data exfiltration requests, secrets and API keys, personal data, toxicity. A la carte: pick the checks (price drops with them), the sensitivity (strict, balanced, lenient) and an optional consensus of independent

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.

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.

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).

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.

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

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.

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.

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.

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