Docs
Use TwinEthos in Claude, ChatGPT and other AI assistants
TwinEthos Lite is free, with no sign-up: ask your AI assistant which AI laws apply to what you're building, and it answers from TwinEthos, citing the exact provision.
What Lite covers (free)
- Which AI laws, regulations and standards apply to an AI feature in a market
- The exact provision, with its official link and its page in the explorer
- Who each duty applies to, and when it applies
- The guard to add, in one sentence, and the few guards that cover the most rules
- What comes into force next
What needs an API key
- Code detectors your coding agent runs against your repository
- The full review checklist, with every detector and how to report each finding
- Bulk data: every rule, guard and source, without the answer caps
- The coding-agent workflow: review a repository and cite file and line
Request MCP access for a key.
Add it to your assistant
The server address:
https://api.twinethos.com/lite/mcp- Claude (claude.ai or the desktop app): open Settings, then Connectors, and add a custom connector named TwinEthos Lite with the address above. It needs no authentication.
- ChatGPT: with developer mode on (Settings, Apps and connectors, Advanced), create a connector with the address above and no authentication. Availability depends on your ChatGPT plan.
- Other MCP clients: any client that supports remote MCP servers over streamable HTTP.
Your assistant decides when to use TwinEthos and writes the answer in its own words from what TwinEthos returns. Coding agents (Claude Code, Cursor, VS Code with GitHub Copilot) get the full server with an API key: see the quickstart.
Questions to try
- Which AI laws apply to a customer-support chatbot used in California and the EU?
- What does the EU AI Act require of a chatbot, and from when?
- What AI rules come into force in the next six months for hiring tools in the US?
Limits
Each answer lists at most 10 rules, or 5 guards. Calls are limited to 20 an hour and 160 a day from one address. Claude's and ChatGPT's users reach TwinEthos from those companies' own servers, so each assistant's users share a larger allowance, and there is a daily limit for everyone. Over a limit, the assistant is told when to try again.
Privacy
TwinEthos receives only what the assistant sends to its tools: market codes, AI feature types, a rule id or a few search words. It never receives your conversation. Each request is logged without its content: the time, the tool, the outcome and whether it came from Claude, ChatGPT or elsewhere; never an address, the tool's inputs or the answer. Your address is used only in memory, to apply the limits. See the terms and notice.
Informational data, not legal advice. Rules have not been reviewed by a lawyer; TwinEthos recommended guardrails are its opinion, not law.