Every agent-payable API.

Multi-list sanctions bundle for agents: OFAC address screen + entity name screen (OFAC/EU/UN index) + optional domain trust + LION_SIGNED_COMPLIANCE_RECEIPT_V1. $0.02 Base USDC, keyless, Ed25519-attested.

Point of interest / business search (keyless, OSM)

LLM chat completion (local RTX 5090 models first, hosted allowlist second) — Genesis402 / UnyKorn Operator Network

Every model on the OpenRouter marketplace with per-million-token input and output pricing, cached-input rates where published, and context windows: several hundred rows, rebuilt daily. Each build carries a changes block diffing it against the previous published build, listing models added, models the marketplace dropped with their last known figures, and every price or context-window movement as from/to. Spot repricing without re-reading 400 unchanged rows.

LLM chat completion per call — no account, no API key, no token math. Three flat-priced tiers: fast $0.002 (DeepSeek v4 Flash), smart $0.02 (GPT-5.4 mini), reasoning $0.03 (DeepSeek v4 Pro). Send OpenAI-style messages, get the assistant reply with finish_reason and token usage. Input capped per tier (16k-32k chars); the 402 quotes the exact tier price up front. Model or source unavailable means a 503 and you pay nothing. USDC on Base.

LLM cost scout: send your monthly OpenAI API spend in USD and get back the flat ChatGPT/Codex subscription tier that absorbs it through the ProxyLLM gateway, the exact new monthly total and savings, the live model catalog, and machine-readable signup steps an agent can complete autonomously (email OTP, no browser).

LLM inference pay-per-call, no account: prompt in, completion out (fast frontier-lab mini model, up to 2000 output tokens). Body: {prompt, system?, max_tokens?}

A verified, machine-readable table of current LLM API pricing (input/output per MTok, cached-input where published) and context windows across major providers (Anthropic, OpenAI, Google, DeepSeek, xAI). Every row carries its official source URL and verification date. Curated and re-verified weekly by the operator agent.

Web search that returns READY-TO-USE grounding text, not links: for one query you get the passages an independent search index extracted from the pages it ranked, flattened into chunks with their source URL and title and an approximate token count, under a token budget you choose (1024-32768, default 4096). Use it to ground a model or a RAG pipeline in one call, where search + extract would take several.

Evaluate whether a requested USDC supply amount is realizable across approved Aave V3 and Morpho Blue lending opportunities on Base, with optional minimum supply APY and protocol filters.

Surfaces the LP holders, how much of the LP supply is locked or burned vs pullable, the lockers, and unlock dates. Unlocked liquidity is the clearest rug setup — this shows the detail a one-line 'LP locked: yes/no' hides.

Social Post / Price Timing Check. Whether social posts led or lagged the price move, measured on Robinhood-chain token launches only — not BTC, ETH or listed majors. Post times are recovered arithmetically from X snowflake ids. The sample is small, outcome-selected from pumped launches, and can be weeks old; its age in days ships with the result. In our sample posts lagged the move, so this is a timing disclosure and never an early-warning or entry signal.

Launch Evidence Digest. Descriptive digest of token launches observed in a time window: count, median terminal multiple, share that ended below -70%/-90%, and the launchpads involved. Facts already observed on chain — not a forecast and not a trading signal.

Sealed Forecast History. Sealed 5-minute BTC forecast rounds: each carries a cutoff timestamp, a feature-snapshot hash, the settlement price source, and multiple agents' submissions. Because rounds are sealed before resolution they cannot be backfilled. A dataset for benchmarking, not a trading signal.

LPC Collection Facts. Collection-level facts for the LPC NFT collection on Polygon: supply, holder count, tier distribution, concentration and marketplace activity. Every response carries the scan time, so the age of the snapshot is never hidden. Aggregates only — no owner addresses, no account-linked data, and not a valuation.

x402 Endpoint Payment Check. Check an x402 endpoint's 402 payment-required body before you rely on it. Reports whether the contract is payable, whether it is opted in to the Bazaar catalogue, and whether a buying agent can read its input schema by machine. Every defect carries a JSON evidence path. Reads the body you supply; it does not pay or call the endpoint. A PASS is not a correctness guarantee and not a claim that the service is accurate.

Check an ACP offering's published contract before you buy. Detects listings that declare no JSON deliverable schema, name required fields they never declare, or specify no input schema at all — the defects that make a purchase unverifiable after the fact. Reads the public listing only; nothing is purchased or executed. Deterministic, no model judgment. A PASS means the listing is machine-checkable — it is not a correctness guarantee and says nothing about output quality.

Audit an ACP deliverable against the schema its seller published. Every defect carries a JSON evidence path, so it can be fixed or disputed; findings without a path are not emitted. The grade comes from deterministic checks only and is reproducible. If the seller published no machine-checkable schema, the result says so instead of passing silently. Credential leakage fails closed. Your deliverable is not stored, resold, or used for training. Audit evidence only: not a correctness guarantee.

On-chain address identity label lookup on Base (100k+ curated entity labels). Pay per call with USDC via x402.

Speichert höchstens zehn Labels ausschließlich im Namensraum des Zahlers. Die verifizierte Zahleradresse ist der Eigentümer; Ergebnisse und gespeicherte Daten sind strikt auf diesen Zahler begrenzt.

Labrador — River — Patient retrieval companion for recovering lost ideas and forgotten context.

Call when an agent needs a heuristic language guess for text. No LLM and no network. Exact $0.002 USDC. Prefer unpaid POST /v1/sandbox/lang-detect first.

Detect the language of a text (80+ languages) with ISO 639-1 and 639-3 codes, confidence, a reliability flag and the top candidates. Runs locally, nothing is stored. Params: text (required, up to 20,000 characters), top (number of candidates, default 3).

Language signal audit fetches current public evidence from the supplied target and evaluates the documented criterion. It returns the observation, threshold, verdict, exact evidence source, limitations and a mode-specific action. The check is passive: no login attempt, payload injection, exploit, purchase, form submission or load test is performed.

Language signal decision fetches current public evidence from the supplied target and evaluates the documented criterion. It returns the observation, threshold, verdict, exact evidence source, limitations and a mode-specific action. The check is passive: no login attempt, payload injection, exploit, purchase, form submission or load test is performed.

Language signal repair plan fetches current public evidence from the supplied target and evaluates the documented criterion. It returns the observation, threshold, verdict, exact evidence source, limitations and a mode-specific action. The check is passive: no login attempt, payload injection, exploit, purchase, form submission or load test is performed.

Read recent large ERC-20 transfers directly from Base, Ethereum, or Arbitrum by scanning Transfer logs over the most recent blocks and filtering by a minimum token amount. Defaults to USDC on Base. Public RPC providers cap the scan window, so this covers the most recent blocks rather than a long history, and it does not label exchange wallets.

Largest-context-window leaderboard: the 100 AI models with the biggest context windows, ranked descending — how many tokens each model can actually take, its max output tokens, modality support, and what those tokens cost. Compare GPT-5, Claude 5, Gemini 3 Pro, Llama 4 Scout and 2,000+ more on context capacity and price together. Refreshed hourly.

F1 Las Vegas Grand Prix race-weekend trip planner at Las Vegas Strip Circuit, Las Vegas, NV (nearest airport LAS). One call returns the session schedule (practice / qualifying / race) + circuit venue intel (coords, airport, transit tips) + Open-Meteo weather + Duffel flights + Duffel Stays hotels. Travel enrichment activates when a single session is selected. Params: session (practice|qualifying|race|<session id>), origin, passengers, check_in, check_out, guests.

Send money to a debit card via push-to-card. Supports USD (U.S.), EUR (Eurozone), and GBP (U.K.). Returns a redemption URL to enter card details.

Total precipitation in the last 48 hours for the given location(s). | ERROR: Unable to calculate price.

Total precipitation in the last 48 hours for the given location(s). | ERROR: Unable to calculate price.

Last N blocks (default 10): height, hash, size, txcount, time — flat $0.001 per call, USDC on Base, no signup.

Verifiable random numbers with cryptographic epoch references. Wesolowski 2048-bit VDF epoch chain with on-chain MerkleAnchor reference. For unbiased randomness: POST /v1/rng/commit (free), then GET the reveal_url returned in the response. Do not construct reveal paths yourself.

Latest Korean brokerage equity research across all covered companies, newest first — company and KRX ticker, target price, rating, brokerage, analyst, and a one-line English summary of each report. Facts and summaries only; the report PDF stays with its publisher and source_url links back to it. Optional ?limit=20 (max 50).

Compact Daily Monitoring snapshot for the tracked slice.

Complete latest daily Keelbase Signal records and full synthesis.

Skim: x402 ecosystem feed — new services and price changes across the public x402 registries, as structured items. Refreshed on a 15-minute cycle (stale served if a registry is down); poll as often as you like.

Latest Base block: number, timestamp and hash. The cheapest call on the service — useful as a liveness and connectivity check.

Latest Base block for Treasury: Return the latest Base block number and timestamp. Focus: treasury. Uses live Coinbase Exchange or Base RPC data.

Latest Base block number and timestamp.

Latest SEC filings for an equity symbol

Latest Solana blockhash and how long it stays valid — Genesis402 / UnyKorn Operator Network

Latest Stellar ledger — height, close time, fees and throughput — Genesis402 / UnyKorn Operator Network

Latest US federal regulations

Latest direction+magnitude scorings for Polymarket events. Each row scores one (headline, market) pair with a direction (up/down/none), confidence (0-1), signed magnitude in points, composite tier (SKIP/WATCH/TRADE), and current market price.

Latest equity quote from a free, key-less source (Yahoo Finance): price, day open/high/low, previous close, % change, volume and currency. Input a symbol like AAPL. Delayed, $0.005.

How well AI agents will use your API, from its OpenAPI 3.0/3.1 spec: a 0-100 score over 13 rules (operationIds, summaries, descriptions, parameters, enums, examples, JSON bodies, response and error schemas, pagination, auth, idempotency, tool count), every finding, and a suggested fix per operation. POST {"url": "https://…/openapi.json"} or {"spec": "<JSON or YAML>"}.

Uses lattice dimension, log2 determinant and an assumed root-Hermite factor to estimate Gaussian-heuristic and reduced-basis vector lengths in log2 units. It does not execute lattice reduction, validate an LWE instance or provide a cryptographic security level.

Who launched this pump.fun token and what happened to their previous ones: launch count, graduation record, and whether the creator bought their own bonding curve