Every agent-payable API.

Text classification (sentiment analysis, topic tagging, intent detection, moderation labels...) into one of your own labels, with a confidence score, via Claude Haiku 4.5. JSON body: {text: string (max 8000 chars), labels: string[] (2 to 20 labels)}.

Classify text into YOUR labels (single or multi-label) with confidence and reason — Genesis402 / UnyKorn Operator Network

Classify text into user-defined categories. Supports single-label and multi-label. MiniMax M2.7 inference.

Classify a User-Agent by device form factor and return convenience booleans (isBot, isMobile, isTablet, isDesktop, isTouchCapable) for routing decisions.

Classify a company's industry using standard codes. Given a company name, description, or website URL, returns SIC, NAICS, and NACE codes with confidence levels.

Classify a field-name inventory for likely sensitive data without inspecting values. Use it as a bounded preflight or analysis step inside an enterprise agent workflow before data, policy, integration, security, or commercial decisions reach production. Returns deterministic machine-readable JSON for $0.003 USDC via x402 on Base. Execution is first-party, local-only, stateless, memory-only, and has no paid upstream or input retention. Results are advisory; the caller remains responsible for aut…

Classify a signal against optional upper and lower thresholds — call when classifying a signal against upper/lower thresholds. Use on caller-supplied ticks/returns when agents need signal prep without a licensed live market feed (ottoai-class demand, first-party transform). Returns deterministic machine-readable JSON for $0.003 USDC via x402 on Base. Execution is first-party, local-only, stateless, and memory-only with no paid upstream, no input retention, and no claim of live chain tip, web se…

Classify dangerous goods by UN number or substance name. Returns hazard class, packing group, labels, marine pollutant status. 200+ substances.

Classify text into categories. Optionally provide custom categories. Returns primary category, confidence scores, topic keywords, and summary.

Classroom Checklist builder: Turn supplied steps into a numbered checklist. Tailored for classroom workflows.

Classroom Simple project planner: Distribute a short task list across available days. Tailored for classroom workflows.

Classroom Weekday counter: Count weekdays between two dates, inclusive. Tailored for classroom workflows.

Claude pricing table: what every Claude model costs per token right now — Claude 5, Claude Opus, Claude Sonnet, Claude Haiku — from Anthropic and every provider that resells or hosts them (Bedrock, Azure, Vertex, OpenRouter, Fireworks). Input, output, cache and batch USD per 1M tokens, sorted cheapest first, so you can compare Claude inference cost across hosts and pick the cheapest place to run one. Normalized, cross-checked, refreshed hourly.

Clean an email list: pass ?emails=a@b.com,c@d.com (up to 100 in one call) and get a verdict per address plus totals. Each address is checked for usable shape, for MX records read live over DNS, against 25 known disposable-mail domains and against 27 role-account names like info, support and admin. Verdicts are deliverable-domain, role-account, disposable, undeliverable, unknown or invalid, and each one says in words what it means.

Extract clean readable text from caller-supplied HTML (no URL fetch).

Clean, LLM-ready markdown from any URL. JavaScript rendering included. Feed web content directly into agent context without parsing HTML.

Clearsigned — the trust layer with no stake in the outcome. Submit {claim, evidence[]}, receive an Ed25519-SIGNED, evidence-bearing receipt (verify offline against GET /pubkey; every receipt is chained in the public ledger at GET /ledger/verify). Flat price, verdict-independent; refunds only for failure to deliver. Current mode is honestly labeled in each receipt.

Returns the German Weather Service reference-period values for the station nearest a coordinate: monthly means and annual totals for air temperature, precipitation and sunshine duration, plus the mean annual counts of frost days, ice days, summer days and hot days.

Clinical Trial Search (ClinicalTrials.gov). Clinical trial search on ClinicalTrials.gov by condition, drug/sponsor/NCT id, location, status and phase: title, phases, sponsor, interventions, enrollment, dates, eligibility and matching trial sites (facility, city, country).

ClinicalTrials.gov registry lookup by company, drug, or intervention; returns study status, phase, sponsor, enrollment, and completion data. Registry data only, not medical advice or proof of efficacy.

Which companies just opened a new clinical trial or site? Bulk feed over NIH ClinicalTrials.gov interventional, industry-sponsored studies, newest-first by registry-posting date (the R&D / commercial-growth signal). Cursor-paginated up to 1000/page, filterable by sponsor, phase, status, condition, intervention type, site state, or since. Rows carry sponsor, phase, status, conditions, enrollment, dates, site count, and the US states sited. Observational registry records, never a prediction.

Find recently updated ClinicalTrials.gov study identifiers, titles and registry status for a condition, with the registry processing date. No medical conclusion.

Lyme disease public use aggregated data with geography, 2022-2023 (Centers for Disease Control and Prevention): one record by its identifier from the agency's published dataset, with a provenance chain and an Ed25519 attestation on every response.

CISA Known Exploited Vulnerabilities status for one CVE: whether it is on the KEV catalog, the due date, required action and ransomware-campaign flag.

Clink Forge raw data product 'auto-nchs-births-and-general-fertility-rates-united-states-lookup': pay-per-call government data with a provenance chain and an Ed25519 attestation on every response.

Cloudflare Workers AI pricing table: per-token cost of every AI model Cloudflare Workers AI 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 Cloudflare Workers AI and against other providers hosting the same model. Normalized from public sources, cross-checked, refreshed hourly.

Where are corporate insiders buying or selling as a cluster? Search SEC Form 4 clusters — issuers where several distinct officers/directors/10%-owners transacted the same direction in a trailing 30- or 90-day window (buy=accumulation, sell=distribution). Filter by ticker, direction, window, minimum insiders, minimum value, or exchange. Returns issuer ticker/company/exchange, insider count, and total shares/value. Public SEC filings as filed — observational, not advice.

CMS Care Compare quality record by CCN for a snf. Answers: what are the quality ratings and performance measures for this facility? Source: CMS Care Compare.

Writes and runs Python to compute or verify an answer, returning the code, its output, and a final answer.

Package security scanner — CVEs, license check, typosquat detection.

Buy an admin-issued prepaid API code for one GPT model and a token budget

Buy a prepaid API code for one GPT model and a token budget

What CHANGED on the board since your last check — change detection for polling agents. Pass ?since=<hash from the FREE /api/board/state> for the deltas: names entering or leaving the READY pool, per-name trend, light, leadership and hold-band flips, and any regime or permission-ladder move per book (S&P 500, Nasdaq-100, macro). Poll the hash free, buy the diff only when it moves — an unchanged hash is answered free, never billed. Recomputes ~5 min in market hours. Research, not advice.

24/7 crypto day-state board — 81 Robinhood-tradable coins (BTC, ETH, SOL, XRP, DOGE…). Per coin: price, 24h/7d change, trend vs 20/50/200-day, ATR%, RS vs BTC, tape lights, volume z, room to the 20d high, UP/CHOP/DOWN state, ranks — PLUS day-trade grades: GO, FUEL and EFF vs the BTC anchor, SETUP shape and ROOM in bars. Publishes every ~5 minutes around the clock when the market moved. A state map, not a ranking. Research, not advice.

Pre-registered overnight continuation read for the Robinhood-tradable crypto universe: the top-5 session-momentum coins into the 20:00 UTC signal, the BTC 200-day regime gate, and the study receipts. The ranking beat its random null at every fee tier in the cold study; at Robinhood retail fees the studied trade was REFUTED — a state read with its refutation attached, not advice. ~15-min refresh, 24/7.

Which of 81 Robinhood-tradable coins are in a WASHED SETUP right now? One read scans the full universe for the two washed-shape patterns (daily uptrend + washed-out 5-minute tape) and returns every hit with its GO score, fuel multiple vs BTC, grade and room — plus the setup column for all 81 names. Recomputed every ~5 minutes, 24/7. Shapes were backtest-confirmed on equities; on crypto they are forward-graded, disclosed per read. Research, not advice.

Day-trade STATE MAP for agents: intraday multi-timeframe conditions on ~70 liquid names (14 market leaders + AI/semis + software cohorts), refreshed ~5 min in US market hours. Per name: an 18-cell lights grid (price/RSI/MACD x 5m-1D), GO (lights x fuel), FUEL vs the tab's anchor ETF, efficiency, ATR%, 20-day vs SPY, entry-window state, and ROOM to the nearest overhead level in % and bars. A state map, NOT a ranking - fully-lit boards flag ext:true (late). Research, not advice.

Per-symbol HOLD-STATE read — is the published trend state behind a name intact? GET /api/board/hold?sym=NVDA (up to 10, comma-separated, one price) returns each name's state, hold strength 0-100, readiness, maturity-late flag, overnight gap-risk grade and its book's regime, with the mechanical rules states flip on. BTC-USD/ETH-USD return the crypto book's IN|CASH daily-close state. Impersonal publication — never your entry or size. ~5-min refresh in market hours. Research, not advice.

Buy 1 hour of Coil Live for $0.10 with a wallet instead of a subscription — the engine-grade board the live record trades from: per-name structural stop and target levels, risk-reward, conviction, lane and ladder placement, plus the served rulebook. Pay once over x402 and receive a session token that works anywhere the X-License-Key header is read, HTTP and MCP alike. The feed behind it refreshes about every 5 minutes in US market hours. No account, no card, no human. Research, not advice.

Sector and macro ROTATION map for agents: where money is actually going, measured as each ETF's ratio to SPY against its own 50-day. ~50 sector + macro ETFs, each labelled GO (leading), HOLDING (paused), TURNING (money arriving) or OUT (lagging), with distance-to-50d %, 20-day slope % and ratio RSI. Pre-bucketed led_by/paused/turning/lagging lists. States and percentages only, never levels. Refreshed every ~5 min in US market hours. Research, not advice.

OHLC candles for a coin: days 1 (30-min candles), 7 or 14 (4-hour), 30 / 90 / 180 / 365 (daily), in any vs currency. Returns aligned {time, open, high, low, close} bars plus the window's high/low/first/last and % change. The candle sibling of crypto-history, which returns close-only points: use OHLC for technical indicators, range/volatility work and charting. Cached in-process for 60 seconds.

Coin fiyati API — anlik kripto fiyati (USD)

Flip one or more fair coins and get heads or tails, with the tally. Cryptographically random, not Math.random. For tie-breaks and A/B splits an agent has to decide on its own.

Price, market cap and volume series for a coin over an ARBITRARY time window: `from`/`to` as unix seconds or ISO dates, or `lookback` ("6h", "3d", "2w") ending now. Granularity is automatic upstream: 5-minute points for a window of 1 day or less, hourly up to 90 days, daily beyond. The windowed sibling of crypto-history (which counts days back from now): use it to align with a specific event, a filing date or another series.

Full profile of one coin by id: name/symbol, description, categories, contract addresses per platform, links (homepage, explorers, repos, social), hashing algorithm, genesis date, and a market snapshot in your currency: price, rank, market cap, fully diluted valuation, 24h volume, ATH/ATL with dates and distance, circulating/total/max supply, 24h/7d/30d/1y change. One call for the metadata a token-risk or research agent otherwise assembles from five. Cached in-process for 5 minutes.

Resolve a name or ticker to coin ids: full-text search returning matching coins (id, symbol, name, market cap rank, ordered by rank), exchanges (id, name, market type) and categories (id, name), plus NFT collections. `limit` per group 1-50 (default 10). The id resolver for every other coin tool: an agent holding "PEPE" or "render" gets the canonical id here before calling coin-profile, coin-ohlc or coin-history. Cached in-process for 5 minutes.

Price, market cap and 24h volume of a coin on ONE past date (00:00 UTC snapshot), in every currency at once plus your chosen one called out. Pass `date` (YYYY-MM-DD) or `daysAgo` (1-365). The point-in-time sibling of crypto-history (which returns a series): use it to value a position as of a date, compute a since-date return, or check a tax-lot basis. Historical range on this plan is the past 365 days, enforced before the call. Cached in-process for 5 minutes.

Search CoinCap assets by name or symbol (agent-optimized).

Full asset data for one or more slugs (agent-optimized).

Historical price series for an asset slug (capped data points).