All Services

Simulates and audits an unsigned Base mainnet transaction — estimating gas, checking balances and bytecode, decoding token transfers/approvals, and returning a PASS/REVIEW/FAIL verdict — without ever signing or broadcasting.

Compares a single macroeconomic indicator across multiple countries, returning latest values with computed signals: rank, highest, lowest, spread, and mean — each cited to its source.

Returns the latest value of a single macroeconomic indicator for multiple countries side-by-side, with computed signals: rank, highest, lowest, spread, mean, and source citations.

Returns a curated basket of key macroeconomic and development indicators (GDP, inflation, unemployment, population, debt, CO2, etc.) for a single country, with latest values and historical data cited to sources.

Returns a curated basket of 10 key macroeconomic and development indicators (GDP, inflation, unemployment, etc.) for a single country, with latest values, short history, and source citations.

Retrieves a cited time-series for one macroeconomic or development indicator across one or more countries, fusing World Bank and IMF data with source attribution per data point.

Returns a cited macroeconomic or development indicator as a time-series for one or more countries, tagging each value with its source, year, and country ISO3.

Converts an amount between currencies using official European Central Bank reference rates, with support for historical rates back to 1999.

Fetches official US Bureau of Labor Statistics data for up to 5 series (CPI, unemployment, wages, etc.), returning latest value, year-ago value, YoY change, and last 13 observations

Returns comprehensive profit-and-loss analytics for a Solana wallet address, including realized/unrealized PnL, drawdown, hold times, daily PnL curve, and open/closed positions.

Calculates the Ranson 1974 pancreatitis severity score from 11 boolean clinical signs (5 at admission, 6 at 48h) and returns total points plus a mortality risk band

Computes the LRINEC score (Wong 2004) from laboratory values to assess risk of necrotizing fasciitis

Computes the Revised International Staging System (R-ISS) stage for multiple myeloma using beta-2 microglobulin, albumin, LDH status, and cytogenetic risk (Palumbo 2015).

Computes the Ley 2012 GAP (Gender, Age, Physiology) index total score (0–8) and stage (1–3) for idiopathic pulmonary fibrosis prognosis

Maps a natural-language drug, diagnosis, or lab phrase to current RxNorm, ICD-10-CM, or LOINC codes without hallucinating billing or pharmacy codes.

Computes the Wells deep-vein thrombosis pretest probability score from 10 clinical flags, returning both three-tier and two-tier risk bands.

Computes the Hemphill 2001 ICH Score (0–6) from caller-supplied GCS, age, infratentorial origin, ICH volume, and intraventricular hemorrhage status.

Computes the McMillan 2007 modified Glasgow Prognostic Score (mGPS 0–2) from serum CRP (mg/L) and albumin (g/L or g/dL)

Computes the Charlson 1987 original weighted comorbidity index score from 19 caller-supplied clinical flags, returning a total score (0–37) and 1-year mortality band.

Applies the validated Canadian CT Head Rule to determine whether CT imaging is indicated for a minor head injury patient based on high-risk and medium-risk clinical factors.

Applies the Canadian C-Spine Rule to determine whether cervical radiography is indicated for an alert, stable adult blunt-trauma patient with GCS 15

Computes the Lee 1999 Revised Cardiac Risk Index score (0–6) and risk class (I–IV) from six binary clinical factors for perioperative cardiac risk estimation.

Calculates Systemic Inflammatory Response Syndrome (SIRS) score using the classic Bone 1992 ACCP/SCCM four-criterion definition, returning whether SIRS is present (≥2 of 4 criteria met).

Calculates the Child-Pugh (Child-Turcotte-Pugh) score and class (A/B/C) for cirrhosis severity from bilirubin, albumin, INR, encephalopathy grade, and ascites.

Computes the SOFA (Sepsis-related Organ Failure Assessment) score across six organ systems using PaO2/FiO2, platelets, bilirubin, MAP/vasopressors, GCS, and creatinine or urine output.

Computes the original Lindström 2003 Finnish Diabetes Risk Score (FINDRISC) from seven clinical inputs and returns the 0–20 integer score plus a binary at_risk flag (cutoff ≥9).

Computes the Angulo 2007 NAFLD fibrosis score from age, BMI, IFG/diabetes status, AST, ALT, platelets, and albumin to estimate likelihood of advanced hepatic fibrosis.

Computes the Bedogni 2006 Fatty Liver Index (FLI) from BMI, waist circumference, GGT, and triglycerides, returning a 0–100 score with steatosis band classification.

Calculates heart-rate-corrected QT interval (QTc) using either the Bazett or Fridericia formula, returning a deterministic result with citation.

Calculates the ABCD2 score (Johnston 2007) to estimate early stroke risk after a transient ischemic attack (TIA), based on age, blood pressure, clinical features, symptom duration, and diabetes.

Computes the Pulmonary Embolism Severity Index (PESI) score and risk class (I–V) for a patient with confirmed PE, based on Aujesky 2005 criteria.

Computes the Wells pulmonary embolism probability score from seven discrete clinical flags, returning both three-tier and two-tier risk bands.

Computes the TIMI risk score (0–14) for ST-elevation MI patients from clinical variables including age, comorbidities, hemodynamics, Killip class, ECG findings, and time to treatment.

Computes the Shipp 1993 International Prognostic Index (IPI) score and risk group for aggressive non-Hodgkin lymphoma from five caller-supplied clinical flags.

Evaluates all eight Kline 2004 PERC criteria to determine whether pulmonary embolism can be ruled out in a low-pretest-probability patient without further workup.

Computes the Clinical Pulmonary Infection Score (CPIS, max 12) using Pugin 1991 criteria from six clinical parameters to aid VAP assessment.

Computes the HAS-BLED bleeding risk score for patients with atrial fibrillation from discrete clinical flags

Computes the Glasgow Coma Scale total score from eye, verbal, and motor sub-scores using the specified 1974 or 1976 edition, returning a deterministic published score with citation.

Computes the Emergency Heart Failure Mortality Risk Grade (EHMRG) score and risk group (1–5b) for acute heart failure patients presenting to the ED, based on the Lee 2012 Table 3 integer scoring model.

Computes estimated glomerular filtration rate (eGFR) using the race-free CKD-EPI 2021 creatinine equation from age, sex, and serum creatinine

Computes the THRIVE (Totaled Health Risks in Vascular Events) score from age, NIHSS, and chronic disease factors to estimate 90-day functional outcome and mortality risk after ischemic stroke.

Returns the deterministic Hunt and Hess 1968 subarachnoid hemorrhage surgical-risk grade (1–5) with citation, adjusting one category worse if serious systemic disease is present, capped at grade 5.

Computes the CHA2DS2-VASc stroke-risk score for atrial fibrillation patients from discrete clinical flags (age, sex, CHF, hypertension, diabetes, prior stroke/TIA, vascular disease).

Validates an email address for deliverability, disposability, and risk using DNS/MX lookup, SMTP verification, and typo suggestions

Computes travel duration and/or distance matrices between multiple origin and destination coordinates using Mapbox routing profiles, billed per-call via x402 micropayment.

Returns a bundled snapshot of 6 key macro risk metrics from FRED — VIX, high-yield and investment-grade credit spreads, financial conditions index, US dollar index, and WTI crude oil price — in a single call.

Enriches a company name, domain, or URL into a verified company profile with per-field evidence, confidence scores, and GLEIF LEI registry matching

Fills a user-supplied JSON schema with values extracted from any URL, combining LLM and deterministic canonicalization with per-field source evidence.

Compares two company or entity names using fuzzy matching and normalization to return a confidence score for deduplication and record linkage.

Fuzzy-match two names or addresses and return a 0-100 similarity score with a match decision, using entity-type-aware normalization and multi-component scoring.

Screens an EVM wallet address against the Chainalysis on-chain oracle to determine if it is sanctioned under OFAC, EU, or UN designation lists

Returns a PASS or BLOCK sanctions verdict for a single EVM wallet address using the Chainalysis on-chain oracle (OFAC).

Compiles structured CSRD (ESRS-E1), IFRS S2, SEC Climate, or TNFD sustainability disclosures from company facts and optional GHG inventory data

Converts a Chart.js v4 JSON config into a PNG chart image (line, bar, pie, doughnut, scatter, radar) with no watermark

Captures a full-rendered screenshot of any public URL using headless Chromium, returning PNG or JPEG with support for full-page, retina, and CSS-selector element capture.

Returns live gas data for Ethereum, Base, Arbitrum, Optimism, and Polygon — including base fee, priority fee tiers (slow/standard/fast), and USD transfer cost pulled directly from chain RPC.

Checks whether a drug carries an FDA boxed warning by scanning multiple manufacturer labels and returns the warning text and label version if found.

Searches the FDA medical device recall database by product name, manufacturer, or model and returns matching recalls with classification, reason, recalling firm, dates, status, and verbatim product descriptions including lot numbers and serial ranges.

Semantically searches an AI agent's private, wallet-scoped persistent memory store and returns the top-k most relevant stored memories.

Fetches a URL and returns clean, LLM-ready markdown (boilerplate/nav/ads stripped) plus a cryptographically signed provenance receipt pinning the markdown to its source.

Returns top-N holder analysis for an ERC-20 token including concentration percentages, contract/EOA flags, and well-known wallet labels for rug/sniper risk assessment.

Returns a median USD price for any ERC-20 token or ticker across CoinGecko, DexScreener, and GeckoTerminal, with per-source quotes and spread in basis points to detect stale feeds or arbitrage windows.

Returns a live gas snapshot for Ethereum-compatible chains including base fee, p50/p90 priority fees in gwei, and USD cost estimates for common transaction types using real-time ETH/USD pricing.

Probes multiple public RPC nodes in parallel for a given EVM chain and returns a 0-100 health score, median block number, RPC lag, head age, and per-node latency.

Aggregates live perpetual funding rates across Hyperliquid, dYdX, and Aevo for a given ticker, returning per-venue 8h funding, annualized APR, open interest, mark price, and cross-venue median APR.

Returns a paginated list of crypto projects with founding metadata, sector, category, and fundraising history, filterable by tags, sector, category, and founding date range.

Returns a ranked list of top cryptocurrency assets that have gained the most mindshare over the past 24 hours, reflecting trending attention and momentum.

Returns organizations and individuals who manage crypto investment funds, with filtering by manager ID and date/amount ranges.

Returns paginated crypto funding rounds with details on investors, stage, and raised amounts, with filtering options.

Returns multi-window return-on-investment (ROI) snapshots for one or more cryptocurrency assets.

Returns paginated cryptocurrency merger and acquisition deals including acquirer, target entity, and transaction values with date filtering.

Returns hourly time-series data for crypto asset social signals including mindshare, sentiment score, and momentum score.

Returns daily historical social signal metrics (mindshare, sentiment, engagement) for X (Twitter) users in the crypto space

Returns daily time-series data of social signals (mindshare, momentum, and sentiment scores) for crypto assets.

Converts a natural language swap intent into a structured crypto swap transaction object ready for execution

Fetches a URL and returns a clean, token-efficient markdown summary with ads, navigation, and scripts stripped out

Extracts rich structured metadata from a URL including Open Graph, Twitter cards, JSON-LD, headings, images, response headers, and more

Runs a structured technical SEO audit on a single web page, producing pass/warn/fail findings across metadata, headings, links, social tags, schema markup, and more.

Generates a Brazilian Pix payment BR Code ('Copia e Cola' string) and QR PNG image per Central Bank spec, enabling instant payment receipt via any Brazilian bank app.

Renders a JavaScript-heavy or noisy webpage with a headless browser and returns its main readable content as clean Markdown

Converts text to high-definition audio using the MiniMax speech-02-hd model with configurable voice, emotion, pitch, speed, and language.

Returns a comprehensive safety dossier for a specific MSHA mining operator, including aggregate violation counts, penalty totals, mine list, fatality events, and identity details.

Converts one or more text strings into dense vector embeddings using Mistral's embedding models (mistral-embed or codestral-embed), paid per-call via x402 micropayment.

Send chat messages to Mistral LLMs (Large, Small, Medium, Codestral, Magistral, Pixtral) via a pay-per-call x402 USDC micropayment gateway

Classifies text for harmful or policy-violating content using Mistral's moderation model, returning safety labels per input string.

Validates LLM output against a JSON Schema (Draft 2020-12 or draft-07), reporting each violation with its JSON path, the broken rule, and a fix suggestion, while also unwrapping code-fenced or prose-wrapped model responses before validation.

Look up normalized metadata for an AI model by name or short alias, returning pricing, context limits, capabilities, and benchmark scores fused from multiple open catalogs.

Look up a single normalized AI model record by name or short alias, returning pricing, context limits, capabilities, and metadata fused from multiple open catalogs.

Returns live per-token pricing data for 2,000+ LLMs across 70+ providers, normalized into a single table with input, output, cache, and batch rates in USD per 1M tokens.

Returns the lowest-cost AI models matching capability constraints (context window, vision, function calling, reasoning) across 2,000+ models and 70+ providers

Returns a comprehensive table of capability specs and constraints for 2,000+ LLMs, including context window size, max output tokens, and feature support flags (vision, audio, function-calling, reasoning).

Returns a ranked cheapest-first list of LLMs that satisfy a given task class and optional capability/price constraints, with a recommended pick and deterministic rationale.

Retrieves recent SEC filing events for specified stock tickers, filterable by form type (8-K, 4, 10-Q, 10-K) and record limit.

Estimates maximum home price, loan amount, and monthly PITI payment based on income, debts, and interest rate using standard agency guidelines

Returns hourly and daily weather forecasts (temperature, precipitation, wind speed, precipitation probability) for any lat/lon worldwide, up to 16 days ahead, with no API key required.

Returns hourly weather forecasts (temperature, precipitation, precipitation probability, wind speed) for any lat/lon worldwide, up to 16 days ahead, with no API key required.

Converts a physical quantity of fuel, electricity, or materials into verified CO2e emissions for a specified country and year, covering 1,000+ activity types.

Fetches a single emission factor record (value, unit, methodology, source, licence) by ID, activity type, BEA code, or spend classification for audit-grade provenance.

Converts a transaction amount and industry classification code (MCC, NAICS, BEA, ISIC, SIC, or NACE) into a verified Scope 3 CO2e emission estimate using licensed emission factors with full provenance across 200+ countries.

Creates time-bounded AI-powered monitoring alerts for DEX and lending protocol metrics across Ethereum, Base, and Arbitrum, triggering a webhook when conditions are met.