All projects matching “check”

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.

Smart contract security check before you approve or pay an address: reads the bytecode, behind proxies too, for privileged functions (mint, pause, blacklist, upgrade), dangerous opcodes and protocol type, and checks whether a single key holds those powers. Adds OFAC SDN screening, who funded the address, sybil-filtered ERC-8004 reputation for agents and a 0-100 score with proceed / hold / block. S

Token honeypot / security risk check before buy or approve. Input ERC20 contract + chain. Returns verdict avoid|caution|ok, is_honeypot, sell/buy tax, blacklist, owner powers. Use for honeypot detection, scam token screening, rug check, token security on Base Ethereum BSC Arbitrum Optimism Polygon.

SECONDED checks EVM transactions, permits, tokens, agent registrations and suspicious messages before you sign or act. Two rival-lab models must agree; NOT VERIFIED means pause or ask a human. Choose a product and submit its exact input. A paid answer is a signed check, not a guarantee.

Email authentication check (SPF, DKIM, DMARC): Check a domain's email authentication and spoofing risk from public DNS: MX and mail provider, SPF (record, all-qualifier, DNS lookup count vs the 10 limit), DMARC policy/pct/rua, DKIM at 30 common selectors, MTA-STS, TLS-RPT, BIMI. Returns grade, spoof

Token risk / token safety check for Base and BSC (BNB Chain) tokens: honeypot detector via live buy/sell simulation (honeypot.is, cross-checked with GoPlus when available), buy tax / sell tax, contract security (mint, pause, blacklist, proxy, owner renounced), holder concentration, liquidity, pool age, LP burn. Returns verdict low/medium/high/honeypot, 0-100 risk score, action and plain reasons. R

DCL Trust Oracle. Call only when the text must include the substring AI and must not include "I cannot be held responsible" or "no guarantees". POST JSON {"response","agent_id"}. COMMIT means those checks passed and the action may proceed. NO_COMMIT means do not execute the checked action. Not a tool-call guardrail, prompt-injection detector, or jailbreak screen; use /evaluate/fast for the phrase

AI and search visibility check for any live URL. Loads the page as Googlebot, Bingbot, OAI-SearchBot, PerplexityBot, GPTBot, ClaudeBot, CCBot and a browser, and flags JSON, empty or blocked pages served to crawlers. Checks robots.txt per bot, sitemap, llms.txt, noindex, text visible without JavaScript, tags, and how many pages Google shows. Returns the top 3 fixes in plain words.

Company verification of a legal entity: whether it exists, is registered, and under which identifier — GLEIF, US SEC EDGAR, the Australian Business Register (ABN, ACN, GST, status), with an ASIC search by ACN or ABN. Relevant wherever a company is named in a contract, invoice or filing: an Australian company check, a company register check, an ABN lookup, or an ABN, ACN, LEI or CIK for a company k

CoverBureau: pay-per-call compliance checking for US certificates of insurance (ACORD 25). Needs NO document: give an NAIC number or carrier name plus a state, and get recognition status, financial-strength rating, and whether it meets your threshold. Use it when you have a carrier but no certificate; validate_coi already checks the carrier when you do. Carrier data is a seed dataset; each respons

Company due diligence in one call (KYB, vendor and supplier checks). Give a company name and country, a registry number, an EU VAT number, an LEI or a website domain. Returns the legal entity from the official register, LEI and parents, VAT validity, a sanctions screen (OFAC, EU, UK) of the company and officers, and the domain's age, mail and TLS, as signed pass/warn/fail checks with sources. Use

Agent-tier preview of the AI-visibility check: live pass/fail scan of the actual website — robots.txt AI-crawler blocks (GPTBot, ClaudeBot, PerplexityBot, Google-Extended + 10 more), llms.txt presence, Schema.org structured data, sitemap presence — plus a deterministic grade. No fix plan, no prioritization, no evidence rows (those are in the $0.50 full check). Any domain, any country.

Is TCP port 9300 (Elasticsearch transport) open on a host? Attempts a connection from Cloudflare's edge and reports open/closed/filtered with the connect time; Cloudflare-fronted hosts are reported as inconclusive because their edge answers for them. Exposure checks for Elasticsearch transport without running a scanner. $0.01 per check.

Cheap pre-purchase recall screen, lite version of buy-check for agents with small per-call budgets. For shopping and buyer agents: one call returns the official CPSC recall screen (government rows verbatim) and a deterministic recall-only verdict, plus a pointer to the full $0.50 buy-check for timing, alternatives, and trust flags.

Before your agent negotiates or shares data with another agent, make both prove they qualify. Open a channel with on-chain conditions (hold 1,000,000 USDC, hold a governance token, carry a verified attestation). Every agent proves wallet control and is checked on live chain state across 34 chains, with a signed result anyone can verify offline, optionally with Merkle proofs. Re-check on demand; wh

Check an open-source package for known vulnerabilities via the OSV database (Google Open Source Vulnerabilities). Pass package name, ecosystem (npm, PyPI, Go, crates.io, Maven, RubyGems, etc.) and optional version. Returns vulnerability IDs, summaries, severity and fixed versions. Essential pre-install check for coding agents.

You get a checked official record: the official source hash and a Bitcoin stamp. That is Official checked records. Compare a model, a tweet, a FRED series, or the wrong table to this hash. If they differ, that copy is dirty. It is $0.25 per record_id. Not a stamp on a free PDF. Agent Trust Score comes from graded Proof of Outcome claims.

Deep verify. Use when a claim is high-stakes, or when Check answer or Extract claims marks it unsure (next_step). Checks it against Wikipedia, Wikidata, Google News, Europe PMC studies, FRED economic data and SEC EDGAR filings, and flags contradiction when sources disagree. Async: returns a job_id; poll GET /v1/jobs/{id} free for verdict, evidence for and against, citations. $0.75 USDC.

Pre-payment honesty screen for any x402 endpoint. Run BEFORE paying: probes the target URL unpaid, parses its 402 challenge, cross-checks price against its /.well-known/x402 manifest and the CDP Bazaar listing, and sanity-checks the payTo wallet on-chain. Returns reachable, challenge validity, price, payTo, Bazaar match, risk flags, and a 0-100 honesty score. Public data only; no bulk dataset.

Read-only EVM state API for wallet snapshots, ERC-20 balances, token allowances, contract code, block heights, and bounded eth_call reads across Ethereum, Base, Arbitrum, Optimism, and Polygon. Fixed price: $0.15 USDC on Base for a batch of up to 100 checks. Only submitted checks run. Results are provider-reported RPC data, not cryptographic proofs.

6 email validation checks in one call — syntax (RFC 5322), MX record verification, disposable detection (5,000+ domains), free provider identification, role-based address detection (admin@, info@), and typo suggestion. Returns a 0-1 confidence score with detailed per-check results. POST method protects email addresses from access logs. No API keys, no multiple service integrations needed. Accepts

AI crawler visibility audit: is your website visible to AI crawlers (GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot) so ChatGPT, Claude and Perplexity can read and cite it? Fetched like an AI crawler, without JavaScript: robots.txt, llms.txt, sitemap, soft 404, JSON-LD, and which crawlers are blocked at the edge. $0.01 per check, one payment per check: every rule, a model verdict, a fix with evid

Single-signal check: is this website's pricing machine-readable for AI agents? Detects Offer schema with price/availability and pricing-transparency signals. Returns yes/no plus underlying checks and fixes. Useful for procurement and comparison agents.

Fact-checking with persistent memory — VERITY remembers prior claims your agent has checked and skips repeat lookups, saving tokens across sessions. Returns CURRENT/OUTDATED/DISPUTED/UNVERIFIABLE verdicts with confidence score, sources, and what changed. 0.10 USDC.

Pre-payment check of an x402 API endpoint: discovery files, live 402 terms, comparable USDC prices, latency, prompt-injection flags and public Coinbase CDP Bazaar payment statistics for the host. Heuristic, not a guarantee. $0.002; unreachable targets are free. Online shops: pre_purchase_merchant_trust_check.

Pay Guard: pre-payment allow/warn/block check for x402. Send your planned purchase (domain, max_price_usd, asset/network) plus the 402 challenge (payTo, amount, asset, network, resource). Returns verdict with wallet↔domain verification, price/asset/network checks, lookalike/URL safety, wash flags from Agentropolis, and a free OFAC SDN sanctions screen of payTo (a screen, not a certificate). Determ

Payment security firewall scan for an OUTGOING x402 payment: PII/secret leak detection in payment metadata, nonce replay detection, overpayment detection, prompt-injection-triggered payment analysis (fast + deep tiers), canonical-USDC asset verification, merchant pinning, velocity/spend caps, and counterparty reputation cross-check. Returns allow/flag/block with per-check reasons and an Ed25519-si

Check an unsigned Base mainnet transaction before signing or broadcasting. Simulates with eth_call, estimates gas, checks sender balance and target bytecode, decodes supported token transfers and approvals, flags unlimited approvals and suspicious transaction shapes, and returns PASS, REVIEW, or FAIL. Never signs or broadcasts.

VAPE exploit_check — Contract verification + proxy-swap surface check. Real on-chain/market data, no simulation.

Before your agent negotiates or shares data with another agent, make both prove they qualify. Open a channel with on-chain conditions (a USDC balance, a governance token, a verified attestation). Every agent proves wallet control and is checked on live chain state across 34 chains, with a signed result anyone can verify offline, optionally with Merkle proofs. Re-check on demand; whoever stops qual