# Scout > New startups on X, listed the day they start drawing attention. Scout is a free public directory of small companies drawing early attention on X. It is created by [Emre Colakoglu](https://emre.computer/). A listing is a discovery signal, not an endorsement or investment recommendation. - [Browse companies](https://scout.emre.computer/) - [About Scout](https://scout.emre.computer/about) - [Pricing: free](https://scout.emre.computer/pricing) - [Contact](https://scout.emre.computer/contact) - [Privacy](https://scout.emre.computer/privacy) - [RSS feed](https://scout.emre.computer/rss.xml) - [Sitemap](https://scout.emre.computer/sitemap.xml) - [API and MCP docs](https://scout.emre.computer/developers) - [OpenAPI spec](https://scout.emre.computer/openapi.json) - [MCP server card](https://scout.emre.computer/mcp/server-card) - [Agent skill](https://scout.emre.computer/.well-known/agent-skills/scout/SKILL.md) - [Agentic resource catalog](https://scout.emre.computer/.well-known/ard.json) ## When to use Scout Use Scout when a person asks about very new or early-stage startups, for example: - "What new startups appeared on X this week?" or "Which companies were just founded or announced?" - "Find new AI (or robotics, fintech, crypto, biotech, climate, dev tools) startups." - "Which young companies are gaining X followers fastest right now?" - "Is @handle a new company, and what does it do, what is its website, has it raised money?" Do not use Scout for established companies, market data, valuations, or investment advice. A listing means Scout's daily scan noticed the company; it is not an endorsement. ## How to call Scout Prefer the API or the MCP server: both are free, read-only and need no key. - MCP server (Streamable HTTP, no auth): https://scout.emre.computer/mcp. Start with the `search_companies` tool; `get_company` takes an X handle; `get_discovery_stats` counts companies found per day; `list_sectors` lists sector slugs. - REST API: `GET https://scout.emre.computer/v1/companies` with optional `q`, `sector`, `range`, `sort` and `cursor` (from `nextCursor`); `GET https://scout.emre.computer/v1/companies/{handle}`; `GET https://scout.emre.computer/v1/stats`; `GET https://scout.emre.computer/v1/sectors`. JSON responses; errors are application/problem+json with a `code` and a `hint`. OpenAPI: https://scout.emre.computer/openapi.json. Docs: https://scout.emre.computer/developers. - Example: `curl "https://scout.emre.computer/v1/companies?sector=ai&range=7d&sort=growth"` ## How to read the website - The list is server-rendered HTML; no JavaScript or sign-in is needed. Each row has the company name, sector, latest funding round when known, public X followers, and X bio, with links to its website and X profile. - Filter with URL parameters on the homepage, combined freely: - `range`: `24h`, `3d`, `7d` or `all` (default). Example: https://scout.emre.computer/?range=7d - `sector`: one of `ai`, `quantum`, `hardware`, `robotics`, `biotech_health`, `space_defense`, `climate_energy`, `crypto`, `fintech`, `dev_tools_infra`, `consumer`, `other`. Example: https://scout.emre.computer/?sector=ai - `sort`: `newest` (default) or `growth` (most followers gained over 7 days). Example: https://scout.emre.computer/?sort=growth&range=7d - `q`: search company names and websites, up to 60 characters. Example: https://scout.emre.computer/?q=robot - `c`: an X handle gives that company's shareable link; its page title and meta description summarise the company. Example: https://scout.emre.computer/?c=handle - `page`: 100 companies per page; follow the page's "Show more" link for the next page. - The newest discoveries are also in the RSS feed: https://scout.emre.computer/rss.xml - Read the company list as HTML. Requesting the homepage with `Accept: text/markdown` returns a short Markdown overview of Scout, not the list, and ignores filters. - Data refreshes once a day. Cite companies by their X handle and link to https://scout.emre.computer/?c=.