Every agent-payable API.

Scan an entire public GitHub repo for malicious AI-skill/supply-chain patterns

South African Repo & prime rate (SARB repo rate and prime lending rate, South Africa's benchmark interest rates, with last MPC change date.) Clean cited JSON with official source URL, effective date and freshness flag.

Pay-per-call research brief for AI agents. Pass ?q=topic; returns JSON summary+bullets via Grok. Use for quick USDC-paid outlines on Base (x402). Agents: discover via Bazaar, pay Exact USDC, get structured JSON.

A full Markdown report on a topic you specify, researched on the live web at the time of the request, with every source linked. As long as the topic needs, never cut short. Typically 2-4 minutes; allow 360 s before giving up. POST {"topic":"..."}.

Receive a bounded hiring report with current counts, company and skill statistics, available historical evidence, and coverage.

x402 market intelligence from the full Agentic Market catalog (31k endpoints) plus our liveness scan: where buyers pay, crowded vs underserved niches, top sellers, dead share. Optional ?q= filter.

Full company report ($0.3 USDC, vs $0.40 bought separately). One call returns the company, financials, news and risk packs for a KRX or US ticker: fundamentals and valuation, up to 8 quarters of statements, 14-day news with sentiment, and a 60-day news/filings risk assessment. Sections that have no data are marked unavailable instead of failing the whole report.

Inspect a Python package’s latest release, Python compatibility, declared dependencies, license metadata and PyPI-reported advisories. No code execution.

Reproduzierbare Bootstrap-Streuung. Zieht mit Zurücklegen aus gelieferten Cash-Ergebnissen; Seed zwingend, höchstens 500.000 Ziehungen. Streuung ist keine Prognose. Alle Zahlen, Kurse und Vorgaben liefert der Aufrufer. Keine eigenen Kursdaten, kein Brokerzugriff, keine Ausführung. Rechenwerkzeug, keine Anlageberatung, keine Empfehlung.

Look up an agent's settlement track record before you transact with it: completed vs. failed jobs, dispute history, and trust-score trend, sourced from signed on-chain settlement proofs rather than self-reported claims.

Identify supplied continuity-risk signals without identity, fraud, or wrongdoing claims.

Reputation scoring with linear trust decay: after its first week a score loses 2-5 points a week by risk cluster (none for a sanctions-flagged entity), plus a bonus of up to 5 points for repeat verifications.

Converts one natural-language task into validated OpenAI-compatible JSON request bodies for multiple open-weight models. Uses the optional models list when supplied; otherwise uses all catalogue models compatible with the input. Supports text, visual input, and model-supported reasoning levels. Returns JSON request bodies only and does not execute inference.

Request coverage for a town ($5). Becomes a top-priority demand signal and starts the records agent; you get $5 of call credits for that town once covered. Refund queued after 14 days if not covered. Verified: 100% golden accuracy (78/78), 100% page-cited, UNKNOWN never charged, data as of 2026-10-26.

Request headers echo: exact JSON of what your HTTP client actually sends - debug auth headers, signatures, x402 payments and proxies in seconds

Request metadata echo API for AI agents — returns your IP, user-agent, timestamp and server time, $0.001 USDC on Base. Debug agent HTTP stacks.

Get a price from a real person for a task in the physical world, such as phoning a business or checking something in person. A human quotes a price and delivery time within 24 h; you accept by paying the quote. The fee is refunded if the request is declined.

Ask a real person for a short written answer where human judgment matters: an opinion, a moral or social call, taste, or native-speaker intuition. Send 'question' (and optional 'context'); a human answers within 24 h. Add follow_up_of to continue a conversation with the same person. Not for facts or code. Declined or late requests are refunded.

Ask a real person a multiple-choice question: how a message feels, which option a person would prefer, or whether to go ahead with an action before you take it. Send 'question' and 2-6 'options'; a human picks one within 24 h. Poll the returned status_url. Not for facts or code. Declined or late requests are refunded.

原価要素から総原価・必要売価・値上げ必要額を計算し、9つの原価要素のうち未入力の要素と確認質問を返す（判定は返さない）

Rerank documents against a query over x402 - the Cohere /rerank wire ({query, documents[], top_n} -> results with relevance_score), served by cohere/rerank-v3.5, $0.002 per call in USDC, no API key, no signup. Up to 50 documents (1,600 chars each, 40k total) and a 500-char query per call. Deterministic, so a byte-identical repeat within 10 minutes is served FREE from cache (X-Cache: hit; opt out with cache:false).

Rerank API, Cohere-style request shape, pay per request: query and up to 50 documents in, relevance scores best first out (BAAI bge-reranker-base). No account or API key; failed calls are not charged.

Rerank documents by relevance. Order up to 50 documents by how well each answers a query, with a relevance score from 0 to 1 (BAAI bge-reranker-base, a cross-encoder). Send documents as a JSON array or one per line; each is read up to 2,000 characters, 40,000 in all. Returns the original index, score and text, best first; top_n keeps only the best. Also available in the Cohere-style shape at /v1/rerank. Use to pick what a model reads after search or vector retrieval.

query -> up to 10 deduped, rendered sources as one token-budgeted bundle (~8,000 tokens combined). Combines /search (three-provider failover, region=tw Taiwan lane) with /content's rendered-page fetching internally, so you get one call instead of assembling search-then-fetch yourself. Sources are deduped by domain. Requires at least 3 successfully fetched sources to succeed -- returns sources_requested and sources_returned so you can see how many of the candidate set actually came back.

Answer a question from the live web with citations in one call: web search + grounded LLM synthesis. Returns answer, cited sources, confidence, follow-ups.

Resolve a UK legal company name or registration number to a sourced Companies House profile. Returns company status, incorporation, SIC codes, registered office, filing due dates and overdue flags. Use for company research and CRM enrichment. Ambiguous names return candidates without settlement. Does not verify trading websites or provide credit scores.

Multi-source web research and company/topic intelligence: runs a live web search, visits the top sources, and returns ranked results with titles, URLs, and content snippets. One call replaces a search-then-scrape pipeline for deep research agents.

Ticker research card for one stock perp in a single JSON: spot quote and US market session (pre-market/regular/after-hours/closed), perp prices on trade.xyz (Hyperliquid xyz), Aster and Lighter, gap % per venue vs spot and vs the Solana xStock, funding and basis, recent gap history (last 72h) and the rule-based long/short/watch/avoid verdict. Data, not financial advice.

Research brief with citations, key points, and caveats from the public web. When to use: short cited answers for agent questions (fair-use snippets only — no full-text dumps). USDC on Base via x402 (cost-floored) — no API key.

Produce a source-backed research brief for an agent task using current web and authoritative sources: send a topic or question and get a brief with the sources and citations it relies on.

Source keyword register for research hypothesis wording with verified-note crosswalk; literal vocabulary only. Reuses exactly one existing qualified meeting-notes request. Supplied source goes to the existing Groq model in the USA; existing processor disclosure and cost/margin retained. Notes quotes/owners/deadlines are verified by the unchanged motor first. Extra rules are deterministic; no semantic deadline resolution or specialist advice.

Academic paper search across Google Scholar, arXiv, Crossref, OpenAlex and Semantic Scholar in one call, merged by DOI / arXiv id / title: title, authors, year, venue, citation count, DOI, abstract, landing page and open-access PDF link. Ranked by relevance, citations and recency; yearFrom and openAccessOnly filters. Half the price of other paper search APIs.

Search published research papers and articles across all publishers (Crossref): title, authors, journal, year, DOI, citation count and abstract, by relevance, date or citations. Use to find and cite sources.

Timestamped source search for publication planning passages; literal vocabulary only. Reuses exactly one existing qualified transcript-package request. Source audio leaves the server for the existing Deepgram processor; existing provider disclosure and cost/margin retained. Deterministic timestamps/text processing only; no acoustic VAD, speaker identity or specialist accuracy claims. Base motor quote plus0.001USDC; article fails before settlement if the motor fails.

Research sample recovery - Base-100 series: Base-100 series for Research sample recovery. Up to 200 supplied typed records. Local descriptive analytics; no enrichment or forecast. One of 25 new profiles across 20 shared operations in this industry. 0.01 USDC includes one bounded base-100 series report.

Multi-source research report with academic-grade sourcing — every claim carries a resolvable citation, not a bare assertion. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Broader source pool than content.youtube-research (web + academic vs. video-only); the deep-dive tier above research.topic-news-scanner's fast scan.

Fast real-time news scan across 44+ curated domains (configs/search_domain_profiles.json), every item source-attributed. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). The quick-scan tier below research.topic-deep-researcher's slower, citation-grade report.

Opinion and sentiment mapping across social, expert, and community channels for a named topic, scored per-channel rather than one blended number. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Same methodology as oracle.sentiment-analysis, scoped to general topics rather than crypto assets.

Research note - Heading outline: Heading outline for Research note. Supplied Markdown or plain text, up to 20,000 characters. Deterministic processing and document templates; no generative prose or fact checking. One of 25 profiles or calculations across 20 shared operations in this sector. 0.01 USDC includes one bounded heading outline deliverable.

Research note - Section template: Section template for Research note. Supplied Markdown or plain text, up to 20,000 characters. Deterministic processing and document templates; no generative prose or fact checking. One of 25 profiles or calculations across 20 shared operations in this sector. 0.01 USDC includes one bounded section template deliverable.

Search scholarly literature across journals, conferences, preprints and books: title, authors, year, venue, citation count, DOI link and abstract where available. Optional fromYear filter and up to 20 results per call. Use for research agents, literature reviews, fact-checking a claim against published work, and finding the DOI for a citation.

Reserve TTL object slot

Resilience Training — Stress test simulation for building capacity under sustained pressure.

Fuzzy-dedupe a list of records into clusters of likely-duplicate entities. Blocks by normalized token prefix, scores with Jaro-Winkler + token-set matching (exact on email/phone), unions matches above threshold, and returns a merged canonical record per cluster with a confidence score. Deterministic, no LLM calls.

Resolve an ENS name (vitalik.eth) or Basename (jesse.base.eth) to an address, or an address to its verified primary ENS name. On-chain, deterministic, no LLM.

Find a listed Korean company by Korean or English name, or by stock code: DART corp_code, Korean and English names and stock code, with candidates when the name is ambiguous.

Resolve an ENS name to an Ethereum address — Genesis402 / UnyKorn Operator Network

Resolves a ticker, CIK or company name (fuzzy, top 5) to 10-digit CIK, tickers, exchange, legal name, SIC code and description. Data comes from SEC EDGAR.

Resolve a JP or GB entity using projected official evidence from NTA/gBizINFO or Companies House; raw upstream responses are not returned. Hard identity conflicts are never verified. / 日本または英国の企業を公式根拠の加工済み投影で確認します。元データは返さず、識別子の矛盾はverifiedになりません。

Resolve DNS records for a hostname — A, AAAA, CNAME, MX, TXT, NS, SOA, CAA or SRV. NXDOMAIN or an absent record type returns an empty record set.