Help me set up Zero — a free tool that lets you find and use extra services you don't have built in (image/video generation, search live social media, or hosting a free webpage). …To setup manually, read Installation Guides
Probes an x402 payment-gated URL and returns a trust score (0–100) with details on price, asset, network, payTo address, response time, and Bazaar listing match
Resolves an ENS name to its Ethereum address (with reverse-check) or an Ethereum address to its primary ENS name, using the ENS Universal Resolver on mainnet including CCIP-Read offchain names.
Returns the root tileset JSON descriptor for Google Maps 3D tiles, enabling initialization of a 3D tile streaming session.
Search for EVM tokens by address, symbol, or name fragment across 13+ chains, returning paginated token metadata from live DEX pool data
Fetches on-chain token metadata (name, symbol, decimals, storage slot) for a given EVM token address on a specified chain
Analyzes wellness or marketing copy and returns an FTC compliance lint report flagging unqualified efficacy claims, disease treatment claims, weight loss hype, and unscientific buzzwords.
Analyzes wellness or marketing copy for FTC compliance issues, flagging unqualified efficacy claims, disease treatment claims, weight loss hype, and unscientific buzzwords.
Analyzes a crypto token contract on multiple chains and returns a LOW_RISK/CAUTION/HIGH_RISK verdict with a 0-100 safety score, sellability, tax, liquidity, owner control, and top three risks.
Returns a human-readable rug risk explanation, score, flags, and verdict for a previously scanned smart contract, given a scan_id from /v1/scan.
Aggregates risk scores from multiple independent Solana token risk advisors for a given mint address, returning a consensus verdict and RugShield-specific flags
Returns risk flags and a verdict for a given Solana token address, including deployer stats and rug pull indicators
Checks a Solana pump.fun token mint address for rug pull risk flags using precision-calibrated advisory signals
Executes a specific agent by ID on the Auto.exchange marketplace, performing tasks like coding, design, or writing with per-request stablecoin billing
Returns a full tokenized real-world asset (RWA) product profile including identity, issuer, TVL, yield/APY, chain distribution, growth trends, and on-chain supply cross-check for a given slug, token symbol, or issuer name.
Performs pre-payment security checks on x402 payment targets and returns an explainable ALLOW/REVIEW/BLOCK/UNKNOWN verdict with risk signals before an AI agent sends funds.
Returns a content-diffed delta feed of OSHA enforcement changes for a specific employer over the last N days, filtering out DOL timestamp churn to surface only genuine changes like new inspections, violations, and penalty adjustments.
Searches CPSC and FDA recall databases by brand or product name, returning hazard details, FDA class, remedy, status, and a deterministic risk score in one unified JSON response.
Performs static security analysis of a public GitHub repository to detect supply-chain threats like exfiltration sinks, obfuscated code execution, credential reads, and install-time curl|bash before an agent installs or runs it.
Fetches a public URL and returns its full page content as clean, LLM-ready Markdown, stripping navigation, ads, and scripts while preserving headings, links, and lists.
Enriches a domain or URL into structured company intelligence including identity, tech stack, social/contact surface, DNS/email infrastructure, and AI-search-readiness score (0-100)
Fetches a public URL and returns a fully structured data bundle including title, description, full text, JSON-LD, OpenGraph/Twitter meta, headings, links, and AI-crawler signals in a single call.
Checks whether a cryptocurrency address appears on official public sanctions lists (OFAC SDN, UK, EU, UN) and returns the matched record details.
Returns deterministic multi-hazard risk scores (wildfire, earthquake, flood, tropical cyclone, volcano, drought) for any geographic coordinates without generating satellite imagery.
Fetches a true-color satellite image (MODIS/VIIRS) for a geographic area and date, returning a base64-encoded JPEG with bounding box metadata.
Diagnoses a stuck Bitcoin transaction and returns a concrete rescue plan (RBF fee bump or CPFP) with recommended fee rates and estimated cost
Retrieves comprehensive rare disease reference data from Orphanet by disease name or ORPHAcode, including definitions, synonyms, cross-references, inheritance patterns, prevalence, and HPO phenotypes.
Returns clinical significance, review status, conditions, genes, and molecular details for a genetic variant queried by rsID, ClinVar VCV ID, or HGVS notation, sourced from NCBI ClinVar.
Looks up detailed genetic condition information from MedlinePlus Genetics (NLM) by condition name, including description, frequency, causes, inheritance pattern, related genes, synonyms, and identifiers.
Returns a verified NASA satellite image URL (true color VIIRS/MODIS or false color) for any geographic location, date, and configurable width.
Performs a fast (~350ms) risk assessment of any EVM wallet address or ENS name, returning a toxic score, risk level, and fired risk flags for pre-transaction screening.
Scans an EVM wallet address or ENS name for on-chain risk signals and returns a toxic score, risk level, and flagged risk categories for off-ramp compliance decisions.
Scans source code or configuration text for hardcoded secrets such as API keys, tokens, private keys, and JWTs, returning findings with severity, location, entropy, and masked excerpts.
Check up to 50 CVE IDs against CISA's Known Exploited Vulnerabilities catalog, returning membership status, vendor/product info, exploitation evidence, ransomware association, and remediation guidance
Detects and locates personally identifiable information (emails, phone numbers, credit cards, IBANs, IPs, SSNs, JWTs) in text, returning typed matches with character offsets and per-type counts
Returns parsed Schedule 13D and 13G beneficial ownership filings from SEC EDGAR for a US-listed company, including reporting persons, percent of class owned, voting/dispositive power, and activist intent.
Disambiguates a researcher by name or ORCID and returns their profile including works count, citation count, h-index, institutional affiliations with ROR IDs, top concepts, and top works.
Returns the full citation graph for an academic paper — works it references (outgoing) and works that cite it (incoming) — along with Semantic Scholar influential-citation counts, given any standard paper identifier or title.
Given a DOI, returns whether a free open-access PDF exists and provides the best PDF link with its license, version, and host type via Unpaywall and OpenAlex.
Resolves a research paper by any identifier (DOI, arXiv, PMID, PMCID, OpenAlex ID, or title) and returns fused canonical metadata from OpenAlex, Crossref, Unpaywall, and Semantic Scholar in a single JSON response.
Resolves a research paper by any identifier (DOI, arXiv, PMID, PMCID, OpenAlex ID, or title) and returns fused canonical metadata from OpenAlex, Crossref, Unpaywall, and Semantic Scholar in a single JSON response.
Search academic papers across OpenAlex, arXiv, and PubMed by keyword query with optional filters for source, year, and open access status
Returns an evidence-backed, side-by-side comparison of two companies, products, technologies, or topics across user-defined criteria with cited sources.
Scores company domains (or discovers in-market companies) using a multi-signal 0–100 buying intent score fusing hiring velocity, funding recency, technographics, and competitor displacement.
Scrapes a public web page and returns its full content converted to Markdown format
Compares two web pages (or a URL against a saved snapshot) and returns a structured diff of content changes, using x402 micropayment authentication
Extracts clean text, links, or HTML from any URL via a pay-per-use scraping API for $0.01 USDC per call on Base.
Independently re-fetches a source URL and returns a cryptographically signed pass/fail/unverifiable verdict confirming whether a scraped claim actually appears on that page.
Fetches a list of videos (or shorts/streams) from a YouTube channel by handle, URL, or channel ID
Search YouTube videos by keyword and return ranked results with metadata including view counts, durations, channel info, and descriptions
Returns WHOIS/RDAP registration data for a domain including registrar, registration/expiry dates, nameservers, DNSSEC status, and EPP status codes
Performs live DNS record lookups via DNS-over-HTTPS for any domain or hostname, returning structured DNS records with DNSSEC validation status
Retrieves subdomains for a given domain by querying Certificate Transparency logs via crt.sh, returning subdomain names, issuers, wildcard status, and certificate date ranges.
Audits a domain's email security configuration including SPF, DKIM, DMARC, MTA-STS, BIMI, and MX records, returning a letter grade and scored report
Performs a live SSL/TLS certificate inspection for a given hostname, returning certificate chain, validity, cipher, protocol, SANs, and days remaining
Checks the registration availability of up to 10 domain names in a single call using RDAP lookup
Returns native ETH and ERC-20 token balances with USD values for a wallet address on Base or Ethereum
Fetches on-chain transaction details for a given hash on Base or Ethereum, including status, gas, value, and decoded ERC-20 token transfers
Fetches on-chain ERC-20 token metadata and current USD price for a given contract address on Base or Ethereum
Resolves ENS names or Basenames to Ethereum addresses (forward lookup) or reverse-resolves addresses to names (reverse lookup) by reading directly from chain state
Returns real-time Chainlink oracle USD prices for up to 10 crypto, stablecoin, FX, or metals symbols from Base or Ethereum chain state
Returns the current USD price of an ERC-20 token by querying Uniswap v3 liquidity pools on Base or Ethereum
Fetches and normalizes RSS, Atom, or JSON feeds (or auto-discovers them from a page URL) into a structured list of items
Fetches Open Graph, Twitter Card, favicon, JSON-LD, and other SEO/link-preview metadata from any public URL
Fetches all hyperlinks from any public web page and returns them as a structured list with URL, anchor text, internal/external classification, and occurrence count.
Full-text search across Wikipedia in any language edition, returning article summaries with attribution metadata
Captures a webpage screenshot and runs a vision model to produce a natural-language description or analysis of the page content.
Removes personally identifiable information (PII) from text using context-aware redaction across medical, legal, financial, and criminal domains
Preflights a Solana memecoin mint address for rug/scam risk signals including holder concentration, dev-wallet activity, sell pressure, liquidity context, and creator-network links
Searches Instagram Reels by keyword or query and returns matching reel results
Search for and retrieve profiles of founders who are currently in stealth or who have exited stealth mode
Search YouTube for videos, channels, or playlists by keyword query
Search Instagram posts by keyword and return matching post results with pagination support
Search for companies using natural language queries or AI-generated structured filters via the Tomba API
Search Facebook posts by keyword and return matching post results with pagination support
Searches Reddit subreddits by keyword or topic query and returns matching subreddit results
Searches across multiple data sources using natural language to find contacts, companies, and enrichment data
Search Reddit posts by keyword, returning paginated results scraped from Reddit
Search and filter M&A events by acquirer, target company, deal amount, geography, industry, and time window, returning structured deal records billed per result.
Search a database of 10M+ companies by name, industry, NAICS code, keywords, location, and employee count, returning paginated company profiles.
Search and filter advertisements run by companies across platforms like LinkedIn, Facebook, Google, TikTok, and Apple, with filters for format, date range, and headlines.
Find companies similar to a given reference company, ranked by similarity score, using a domain, LinkedIn URL, or other identifier.
Search real-time startup funding events by round type, amount, geography, industry, investor, and date window, returning structured deal records billed per result.
Search Facebook for group profiles matching given keywords, returning paginated group data
Searches a daily-crawled global index of open job postings, filterable by title, company, location, seniority, team, industry, and recency, billed per result returned.
Searches academic papers and research articles using OpenAlex, returning ranked results with title, authors, year, venue, DOI, citation count, and open-access status.
Fetches live Google Images results for a query, returning ranked images with title, direct image URL, source page URL, and alt text, localised by location and language.
Returns live organic web search results (title, URL, domain, snippet) from Google or Bing for a given query, localized by location and language.
Fetches live, localised Google News results for a query, returning ranked articles with title, URL, domain, source, and timestamp.
Returns a compiled federal OSHA safety dossier for a specific employer (Walt Disney Parks and Resorts US Inc), including severe injury reports, Form 300/301 filings, inspection records, citation details, and trend flags.
Search for Facebook page profiles by keyword, returning matching page results with pagination support
Captures a screenshot of any URL and returns a PNG or JPEG image, paid via crypto micropayment with no logging or tracking
Searches for TikTok users by a text query, returning matching user profiles.
Search the World Bank's catalog of 1,400+ indicators by keyword, returning indicator codes for use in downstream data queries.
Checks whether a Seattle rental property is registered and actively compliant, returning registration status, unit count, expiration date, and any violation cases for a given address.
Fetches and diffs consecutive 13F-HR quarterly filings for an institutional manager to surface opened, closed, increased, and decreased positions with dollar values and portfolio weights.
Retrieves a time series of XBRL financial concept values (e.g. NetIncomeLoss, Revenue) for a given company CIK from SEC EDGAR filings
Looks up SEC EDGAR company identifiers (CIK) and basic profile by ticker symbol or company name
Retrieves SEC EDGAR company profile, recent filings, and XBRL financial fundamentals by ticker symbol or CIK, paid per call in USDC via x402
Retrieves recent SEC filings (10-K, 10-Q, 8-K, Form 4, etc.) for a company by ticker or CIK, with document links, paid per call in USDC via x402
Generates a synthesized, AI-written report of a US company's recent SEC filings from EDGAR, given a ticker or CIK, with full document reading and source citations.