Clawgate sits between your developers and the AI providers. Your engineers keep Claude Code, Codex, and OpenCode exactly as they are — one command points them at your gateway, and every request from that moment on is budgeted, attributed, and recorded.
No new tools for your engineers to learn. It’s a one-line endpoint change.
Point Claude Code, Codex, or OpenCode at your Clawgate gateway and hand each developer a vsk_… key. That’s the whole setup.
Define budgets, allowed/forced models, and per-project rules with simple toggles. No code, no config files.
Track spend per user and project, get monthly invoices, and surface abuse signals before they cost you.
One command points your CLI at Clawgate — it prompts for the developer’s
vsk_… key and writes the config for you.
curl -fsSL https://console.clawgateai.com/config/claude | bashcurl -fsSL https://console.clawgateai.com/config/codex | bashcurl -fsSL https://console.clawgateai.com/config/opencode | bashYour rules are applied before a request ever reaches a model, so a budget is a hard stop rather than a report you read afterwards.
The developer’s vsk_… key identifies the person, and the project tag identifies the work. Every request is attributed before it runs, not reconstructed from an invoice later.
The project’s spend cap is checked first, then the developer’s daily and weekly limits on tokens, dollars, and sessions. Over the line, the request is refused with a clear message instead of being billed.
If the CLI asks for a model you have not allowed, Clawgate serves the best model you have allowed instead of failing. Turn on smart routing and routine work is served by a lighter, cheaper model automatically.
The request goes to the provider, the response streams straight back to the developer, and the exact token counts and cost land on the dashboard. Failed requests are never billed, and cache savings are passed through.