Simple pricing. Your AI, your tokens.
You bring your own AI key and pay your provider for tokens. We charge for the platform: plans differ by projects and runs, and every plan has every agent.
Free
Try the agents on one repository.
- Includes
- 1 project
- 30 runs a month
- 1 run at a time
- Every agent and AI provider
- Website, e-mail and Telegram
- Unlimited members
Pro
For a product team with a few repositories.
- Everything in Free, plus
- 3 projects
- 300 runs a month
- 2 runs at a time
- E-mail support
Team
For companies running agents across products.
- Everything in Pro, plus
- 10 projects
- 1,500 runs a month
- 5 runs at a time
- Slack
- Approval rules per agent
- Audit log export
Enterprise
For larger teams with security and procurement needs.
- Everything in Team, plus
- Unlimited projects and runs
- Runs at a time to fit you
- SSO (SAML)
- Self-hosted in your cloud
- SLA and invoice billing
Compare plans
What each plan includes
| Plan | Free | Pro | Team | Enterprise |
|---|---|---|---|---|
| Usage | ||||
| Projects (repositories) | 1 | 3 | 10 | Unlimited |
| Runs a month | 30 | 300 | 1,500 | Unlimited |
| Runs at a time | 1 | 2 | 5 | Custom |
| Members | Unlimited | Unlimited | Unlimited | Unlimited |
| Agents and AI | ||||
| All eight agents | ||||
| Anthropic, OpenAI, Bedrock, Azure | ||||
| A model per agent, cost per run | ||||
Your own agent prompts and skills (.agents/) | ||||
| Integrations and channels | ||||
| GitHub, GitLab, Jira | ||||
| GA4, PostHog, Sentry, Clarity | ||||
| Website, e-mail, Telegram | ||||
| Slack | – | – | ||
| Control and security | ||||
| Approvals, password for merges, audit log | ||||
| Approval rules per agent | – | – | ||
| Audit log export | – | – | ||
| SSO (SAML) | – | – | – | |
| Self-hosted in your cloud | – | – | – | |
| SLA, invoice billing | – | – | – | |
Questions
Pricing questions
What counts as a run?
One agent job: a product review, a plan, an implementation, a revision, a PR review, a CI fix, a triage, or an answer in the chat. Runs that fail before the agent starts don't count.
Do you charge for AI tokens?
No. Agents use your own key with Anthropic, OpenAI, AWS Bedrock or Azure OpenAI, and your provider bills you for the tokens directly. We show the tokens and the estimated cost of every run, so you always know what an agent costs.
What happens when we use up our runs?
New runs are skipped with a notice until the 1st of the month or until you upgrade. We never charge for extra runs automatically.
Can we use a Claude Pro or Max subscription instead of an API key?
No. Anthropic doesn't allow a service like this one to run on someone's Claude subscription, so agents need an API key from the Anthropic Console, or Claude through AWS Bedrock.
Is VAT included?
Prices exclude VAT, which is added at checkout where it applies. Businesses with a valid VAT ID from another EU country are invoiced under the reverse charge.
Can we cancel or change plans?
Yes, at any time from the billing page. An upgrade starts at once and is prorated; a downgrade or cancellation applies from the next billing period.
Where is our data stored, and is it used for training?
In the United States (Virginia), encrypted at rest. We never train models on your code or data. Your AI provider's own data policy applies to what agents send it.
Can we run it in our own cloud?
Self-hosting is part of the Enterprise plan. Tell us about your setup and we'll come back to you.
Start with the Free plan
One repository, every agent, your own AI key. Ask for early access and we'll invite you soon.