All projects matching “reasoning”

Screen a product, firm, or brand against FDA drug, device, and food recall (enforcement) history. Returns matching recall events with classification, status, and reason. Public-domain openFDA data.

Lists entities publicly flagged as non-compliant under MiCA by EU national competent authorities. Query parameter: country (string, optional; 2-letter EU member-state code, e.g. IT). JSON response: count (number) and entities[] with legal_name, commercial_name, lei, home_member_state, competent_authority, website, infringement, reason, decision_date (ISO date), record_updated. Data source: ESMA In

Spend preflight for AI agents. Send the x402 PaymentRequired body (v1 or v2) you received, or a cart, plus optional rules (per-payment max, hold threshold, daily cap, allowlists, new-domain days, endpoint category price multiple, liveness and payTo changes). Returns allow|hold|block with reasons, the cheapest acceptable payment option, and a receipt id. Read-only: never signs, settles, or holds fu

Is this token safe to buy? Simulates a buy and sell-back through the token's real pool (at your trade size, if given), reads what its creator can still do (mint, hidden mint, pause, upgrade), who can pull the liquidity, the holders, and whether the creator was paid by a wallet behind earlier rugs. A signed allow, caution or block, an action and the reasons.

Fast, cheap spam-risk gate for one email: pass/warn/fail, a 0-100 score, and up to 3 flagged phrases. Use for a go/no-go before sending or for bulk screening of many drafts. No rewrites. For a full audit with reasons, subject-line risk, and rewritten copy, use analyze-email-deliverability. Content signals only; no inbox-placement guarantee.

Know Your Merchant (KYM) trust score: pass a merchant domain and get a deterministic 0-100 trust score with a band (high/medium/low/very_low/avoid) and human-readable reasons. Live signals: domain age from the RDAP registry of record, malware/phishing blocklist verdict via Cloudflare Security DNS, TLS certificate health, mail capability, and refund/privacy/contact page presence. The merchant-side

Call before an agent runs npm install. Checks one npm package against the registry, OSV advisories, install scripts, typosquat and combosquat names and repository provenance, and flags hallucinated or slopsquatted names. Answers safe, caution, block or does_not_exist; every reason is sourced and labelled fact or heuristic.

Weekly seller rates, all published rows, JSON. Realised seller premiums against the ECB reference for every ISO week since the first Base fill, per route and per currency. Route-weeks that fail the publishing rules keep only the reason; a currency-week aggregates only that week's published routes, so it never reveals a withheld one.

Check an x402 seller before you pay it. Probes the endpoint now, checks that its payment can actually succeed (token EIP-712 domain, payTo), and adds its 30-day uptime from our monitor of paid Bazaar listings, how many distinct buyers paid it, and how its price compares with similar listings, with cheaper well-used alternatives. Returns a verdict (ok, caution, avoid) with the reasons.

Assess how trustworthy a domain is using authoritative live signals, not just its spelling. Fuses RDAP registration data (domain age — the single strongest phishing signal — plus registrar, expiry and registry status flags), DNS resolvability (A/AAAA), MX presence and DNSSEC into a 0-100 risk score with a verdict (low/medium/high) and the exact reasons. Pairs with `urltrust` (lexical) for a comple

Verify a New Zealand charity against the Charities Services Charities Register. Look up by charity registration number (CC…) or name; returns registration status verdict (registered/deregistered), registration date, deregistration date and reason where applicable, main sector and activity, and the charity's stated charitable purpose. Register open data licensed CC BY 3.0 NZ.

Verify a JSON Web Token's signature and claims against a shared secret (HS*) or a PEM public key (RS/ES/PS). Returns { valid, payload } on success or { valid: false, reason } for a bad signature, expired/not-yet-active token, or a failed issuer/audience/subject check. Pass `algorithms` to pin the accepted signing algorithms and prevent algorithm-confusion attacks.

Given any public HTTP(S) URL, returns a measured liveness verdict - LIVE, PAUSED, GATED, CHALLENGED, DEGRADED, DEAD or UNKNOWN - with every reason carrying the source it was read from. The point is the case a status check misses: a front door answering 200 over a back end that stopped. Also sold at this origin: good ($149.00, POST /x402/good) - The x402 rail, ninety days - framed print, 24 x 36 in

Email address check: verdict (deliverable_likely/risky/invalid) with reasons from syntax, MX/A records, disposable-provider list, role address, free provider, typo suggestion, SPF/DMARC/DKIM and domain age. Use when an agent is about to send to, sign up, or accept an email address and wants to avoid bounces, throwaway or mistyped addresses. One address per call; no SMTP mailbox probing.

One page load answers everything an agent needs to reason about a page: the DOM after JavaScript capped at 2 MB, the CDP accessibility tree flattened with parent references, every link with rel and target and an internal or external verdict, every form field with its label and autocomplete hint, console errors and warnings with source and line, a screenshot, navigation timings and a summary of the

Detect material state changes with prior/current comparison, evidence, and provenance, applied to SAM.gov federal contract opportunities. Identify economically material deadline moves, solicitation amendments, attachment changes, set-aside, NAICS, status, and award changes. Returns an immutable before/after delta with materiality, reason codes, and provenance for procurement and capture agents.

Token safety check on Base before buying: honeypot, rug pull and scam detection. Verdict SAFE/CAUTION/SCAM with fixed reason codes and action ALLOW/BLOCK/RECHECK; admin rights, sell tax, liquidity, two independent honeypot sources plus own sale simulation of a real holder (Uniswap v4 included), check coverage and block number

Shareable verdict card for a Base or Solana token: grade BASED/MID/COOKED/RADIOACTIVE, 0-100 score, one-liner, reasons, key market metrics, plus a share URL and a 1200x630 PNG. Body {"tokenAddress":"0x..."} or {"chain":"solana","tokenAddress":"<mint>"}.

Decision arbitration: submit context, options, constraints, and risk_tolerance; receive a structured JSON verdict with confidence score, reasoning, and escalation flag.

Website trust check: verdict (trustworthy_likely/caution/risky) with reasons from domain age (RDAP), DNS (A/MX/NS), HTTPS reachability and redirect chain, security headers, phishing/malware list snapshots, brand lookalike/typosquat detection, tech fingerprint and page title. Use when an agent is about to open, click, submit credentials or payment to, recommend or trust an unfamiliar URL or website

One screened human answers an approval, verification or judgment task: approve or reject, true or false, or a label, with a reason and the responder's disclosed identity. $1.25 at a 24-hour deadline. Answered by hand within the deadline or refunded in full automatically. Refused: CAPTCHAs, logins, third-party contact, medical, legal or financial advice.

Solana token safety check: verdict (safe/caution/avoid), 0-100 risk score and reasons from mint/freeze authority, Token-2022 extensions, holder concentration, age and DEX liquidity. Use when an agent is about to buy, swap, accept or recommend an unfamiliar Solana token. Not financial advice.

Check an x402 seller before you pay it. Probes the endpoint now, checks that its payment can actually succeed (token EIP-712 domain, payTo), and adds its 30-day uptime from our monitor of paid Bazaar listings, how many distinct buyers paid it, and how its price compares with similar listings, with cheaper well-used alternatives. Returns a verdict (ok, caution, avoid) with the reasons.

OSCAL evidence bundle for one obligation: EU AI Act Article 50/53, DORA or EU-CRA. Returns selected card objects with signature fields, any retained inclusion proofs, an OSCAL wrapper and assembly manifest. Does not verify input-card signatures. Reports manifest signing status and any unsigned reason. States: ASSEMBLED or ASSEMBLED-EMPTY; zero selected cards stays zero. Supported-card verification

Check a Base or BNB Chain (?chain=bsc) address before you pay it: a verdict label pay / hold / unknown with a human text (Pay, Hold, Inactive, No track record or Couldn't check) and the one reason, such as active with real counterparties or a high-severity flag named. Under it, the evidence: every flag with its source, what was not assessed, and the EAS attestation if one exists. On-chain behavior

Token verdict (green/orange/red, grade SAFE/CAUTION/RISKY/AVOID, reason codes, one-line summary, market data) for a Solana or EVM token: mint/freeze authority, honeypot, tax, LP lock, liquidity, age, holder concentration

AI text summarization: condenses up to 20k characters into a crisp 250-word summary. Use when you have a long document, article, or transcript and need the gist before reasoning over it.

Validate an intended paid API request against its declared contract before spending. Returns VALID, INVALID, or UNKNOWN with machine-readable reasons. Target URL is never called.

Solana wallet profile: verdict (established/new/suspicious/inactive) with reasons from SOL balance, token holdings, wallet age (first tx), 24h/7d activity, program-account detection, bot-like pattern heuristic and OFAC sanctions match. Use when an agent is about to pay, airdrop to, trade with, trust or whitelist an unfamiliar Solana wallet. Not financial advice.

Offline pre-sign check (no RPC, fast): decodes a transaction, EIP-712 typed data (Permit, Permit2, EIP-3009), an EIP-7702 authorization or an x402 402 + payment payload and returns SAFE/CAUTION/DANGER, every risk flag with its reason, decoded args, counterparties, x402 payment-vs-402 match and OFAC SDN screening. Use /v1/check for on-chain facts and simulation.