The aggregate call is the fast path and reports a confidence score based on how many sources actually answered, which is more honest than a profile that looks complete either way. Use the single-facet tools when you know what you want — a tech-stack read from headers, DNS and GitHub is far cheaper than a full profile — and note which inputs each expects, since some take a domain and others the legal company name. The infrastructure tools need no keys at all, so domain intel and social presence work the same wherever it runs.
A company research aggregator. One call fans out across public sources in parallel and returns a merged profile with a confidence score; the other tools go after one facet each when that is all you need.
- A full profile from a name or a domain, aggregating Wikipedia, GitHub, SEC EDGAR, OpenCorporates, RDAP, DNS, web scraping, USPTO patents, news and careers pages in parallel
- The technology stack inferred from HTTP headers, DNS records and GitHub repositories — frameworks, languages, hosting providers, analytics and CDNs
- Key people from the company's own about and team pages, cross-referenced against Wikipedia and the GitHub organisation, returned with titles and sources
- Corporate registry data from OpenCorporates across 140+ jurisdictions: incorporation date, status, jurisdiction, registered address and officers
- SEC EDGAR financials for US public companies — revenue, net income, assets, liabilities, equity, tickers, SIC code and recent 10-K, 10-Q and 8-K filings
- US patents assigned to a company from the USPTO PatentsView API, from 2020 onwards, with numbers, titles, dates, abstracts and inventors
- Domain intelligence with no key at all: A, AAAA, MX, NS, TXT and CNAME records, WHOIS/RDAP registration, and the hosting provider and email service inferred from them
- Open roles scraped from careers pages with titles, departments and locations, including detection of Lever, Greenhouse or Ashby behind them
- Social presence across 12 platforms plus GitHub organisation stats — repositories, stars, languages and followers
- Competitors and recent news, both drawn from web search
Nothing on your side. Locally it runs as npx companyscope-mcp; a hosted Streamable HTTP endpoint on Cloudflare Workers takes no authentication and is capped at 25 calls a day, and a pay-per-use Apify actor exists for heavier use behind a bearer token. Two tools depend on keys held by whoever runs the server rather than by you: competitor discovery needs a Brave Search key configured, and news coverage varies with the same server-side configuration. Patents only go back to 2020, registry lookups want the legal name rather than the trading name, and financials only return for US public companies. MIT.
One command — npx companyscope-mcp