Where do my API keys and questions go when Claude asks ChatGPT and Gemini through Kirat?
Short answer
Everything you give Kirat goes directly from your computer, or from your own Claude Code cloud environment, to OpenAI and Google, under your accounts and their terms. There is no Kirat server, no account with us and no telemetry; we never see your keys or your questions. Your keys stay in your operating system's secure storage or in Claude Code's credential store.
Facts at a glance
- Claude Desktop
- Keys in macOS Keychain or Windows Credential Manager
- Claude Code
- macOS Keychain; on Windows and Linux in
~/.claude/.credentials.json, readable only by your user - Cloud sessions
- Network secrets (unreadable in sessions) or environment variables (readable by everyone using the environment)
- On your computer
- A folder
.kiratwith the cost counter, cost log and the last 20 results - Telemetry
- None
Minimal key permissions
At OpenAI, a restricted key is enough, with “List models” on Read and “Model capabilities” on Request, everything else on None. A Gemini key from Google AI Studio is already limited to the Gemini API; restrict a key from the Google Cloud console to the Generative Language API.
When Kirat is the wrong choice
If your data may not go to OpenAI or Google. Everything you give Kirat goes to both, under your accounts and their terms.
Sources on this site
More questions
- How do I get a second opinion from ChatGPT and Gemini inside Claude?
- What is an AI council, and when is it worth using one?
- Can Claude Code get a code review from GPT and Gemini?
- What does it cost to let ChatGPT, Gemini and Claude vote on a question?
- Does Kirat work in claude.ai, the Claude mobile app or Claude Code on the web?
- Kirat, llm-council or PAL MCP: which multi-model tool fits?
- How can I check an AI answer when I can't tell whether it is right?
Terms in this answer: Network secret