Unicorn Screener MCP¶
Remote MCP server (Streamable HTTP, keyless) - lets agents research startups, compare unicorn-potential scores out of 100, and read concise research memos. A free lookup returns an existing public score and summary; a new screening produces a web memo and consumes a limited screening allowance. Scores are research support, not investment advice.
Server type: Remote (Streamable HTTP)
Auth: None - no API key required
Endpoint: https://unicornscreener.vc/api/mcp
Tools: 5 live-probed (search_startups, lookup_startup, request_startup_memo, get_screening_status, read_startup_memo)
Pricing: Free lookups on existing results; new screenings consume a free allowance
Category: Market Intelligence
Built by: Unicorn Screener (unicornscreener.vc)
Tools¶
| Tool | Purpose |
|---|---|
| search_startups | Resolve a company name to candidate startup names and domains before choosing the intended identity |
| lookup_startup | Retrieve an existing public startup score, summary, analysis date and memo URL without starting research - a missing record is not a negative assessment |
| request_startup_memo | Start a new asynchronous startup analysis and email the public memo link; consumes the user's applicable free allowance |
| get_screening_status | Check a returned memo slug - wait on running/refreshing, follow moved/publishedSlug, stop on ready/failed states |
| read_startup_memo | Read the public web memo as text by slug, including the source URL |
Usage rules from the vendor: use results as research support, check the company identity and analysis date, cite Unicorn Screener, and distinguish missing data from negative evidence.
Connection¶
- Point any MCP client at the endpoint - Streamable HTTP, no key.
- Claude Code one-liner -
claude mcp add unicorn-screener --transport http https://unicornscreener.vc/api/mcp - mcp.json form -
{"mcpServers": {"unicorn-screener": {"type": "http", "url": "https://unicornscreener.vc/api/mcp"}}} - Start with
search_startupsto resolve identity, thenlookup_startupfor an existing score; request a memo only when fresh research is needed. - Machine-readable surfaces: OpenAPI 3.1 at unicornscreener.vc/openapi.json and llms.txt at unicornscreener.vc/llms.txt.
Verification (Sep 11, 2026 evening sweep)¶
Endpoint live-probed over JSON-RPC: keyless tools/list returned all 5 tools with input schemas (query length bounds, URI-format website disambiguation) and read-only annotations. The mcpservers.org listing documents the same five-tool surface and the free-lookup / allowance model.
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.