Use cases · Teams that keep data in-house
Run Token Controller on your own servers.
The same Docker image as the hosted Team plan, with every Team plan feature, on your own Postgres in your own region. Free under the AGPL.
What your security team will ask.
- Where does our cost data live?
- What leaves a person's laptop?
- What do we have to run and keep up to date?
One container, one database.
The service is one program, written in Rust, in one Docker image. All data lives in one Postgres database.
- 1
Start Postgres and the service
With Docker Compose, behind your own reverse proxy with TLS.
- 2
Set your public URL and email
Any SMTP server sends the magic links and reminders for late token sheets.
- 3
Point the local tools at it
Run tc connect with your URL, and send telemetry to your own ingest endpoint.
What stays where.
Transcripts stay on the machine
Prompts, tool output and file contents never leave a laptop. The summary and the tag values are the only text in a token entry.
Open source
The service is AGPL-3.0. Issues on GitHub are open, and community support happens there.
Support when you need it
With a support contract, a self-hosted instance becomes an Enterprise plan instance with OIDC, SCIM and an audit log. If the contract ends, your data and reports stay.
In the docs
Other questions Token Controller answers.
- Agency owners and account managersPass AI agent cost through to your clients
- Controllers and financeClose the month against your provider bills
- Project managersToken budgets per project and ticket
- Engineering managers and heads of departmentA token budget for each person
- Directors of delivery and account managersQuote new work from what similar work cost
- Teams with support, security and other agentsAgents that don't write code
Start alone in a few minutes. Add your team when the first month closes.
No card for the free plan. Fourteen days to try the Team plan.