Which domain officially belongs to which software product or company. Ownership only, never safety.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent β or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag β we're steadily working through the catalog.
π‘ Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
An open registry of which domain belongs to which organization. Every claim is backed by reproducible machine evidence and delivered straight into the paths AI agents actually use β MCP and a plain HTTP API.
Ownership only, never safety. We would rather say "don't know" than be wrong.
Site: https://realurls.org Β· API: https://api.realurls.org Β· MCP: npx -y @realurls/mcp Β· Signed dataset: latest release
δΈζθ―΄ζθ§ docs/zh/README.mdγ
AI assistants and search engines increasingly answer "what is X's official site?" and "where do I download X?". In 2026 that path was attacked in public: an SEO-poisoning campaign pushed fake download sites to the top of search results (the Black Cat campaign infected roughly 278,000 machines), security vendors documented lookalike sites impersonating Claude Code and Gemini CLI, and Microsoft confirmed cases where users asking an LLM for a download link were sent to attacker-controlled domains.
The root cause is simple: there is no open, programmable, evidence-backed source of truth for "which domain is really theirs".
The blacklist side is saturated (PhishTank, Phishing.Database, openSquatβ¦). Those answer "is this domain bad?". Realurls answers the other question: "who does this domain belong to?"
A list says "trust me, this is the official site". We say "here are five independent pieces of evidence, with the commands β run them yourself".
The full trust model is in TRUST.md; the decision rules are in POLICY.md.
First category was AI and developer tools. The registry is now expanding to every software company and open-source project with a real footprint, in reviewed batches. Target: 10,000+ organizations, browsable by category on realurls.org. Precision stays the only hard metric; coverage follows the evidence.
The Build entities from seeds workflow runs the same thing across 16 runners and opens a bot pull request per batch; nothing reaches main before the rules regression and the 200-record manual sample pass.
You contribute leads, not data. Open an issue; a bot collects the evidence and decides whether the record qualifies. See CONTRIBUTING.md.
If you control a domain, a single DNS TXT record (A5) overrides any verdict of ours.
entities/, dist/): CC BY-SA 4.0src/, tests/, api/, mcp/, extension/): MITRealurls judges domain ownership only. It does not judge whether a site is safe, lawful, or any good. A domain marked verified means it really belongs to that organization β not that it is safe. For safety, rely on Google Safe Browsing, VirusTotal and similar services.
No reviews yet β be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/realurls)<a href="https://allmcps.com/mcp/realurls"><img src="https://allmcps.com/api/badge/realurls?style=directory" alt="Realurls on AllMCPs" /></a>