Run CrewAI crews on a governed runtime — Kohala
Give your CrewAI crews charters, spend caps, and cost metering. Import your crew or wrap it with the kohala SDK — every kickoff runs governed, with per-step cost breakdown and a trace ID.
Explore Kohala
Frequently Asked Questions
- Can I run a CrewAI crew on Kohala?
- Yes. Paste your crew code into the import wizard and it runs on Kohala's governed runtime under a recorded charter, a server-enforced allowlist on the Kohala tools it can call, and a billing cap — unscheduled until you give it a cadence. Or wrap the crew with the kohala Python SDK on your own infrastructure and report each kickoff, metered against the same caps and checked by the same output validators. Both paths report credit cost and a trace ID per run.
- Do all agents in my crew share one billing cap?
- The crew runs as one governed agent with a single charter and billing cap, and the run trace breaks down cost across the crew's steps, so you can see exactly where credits went.
- Does Kohala change how my crew's agents collaborate?
- No. CrewAI keeps orchestrating roles, tasks, and delegation. On the hosted runtime Kohala governs the boundary: which Kohala tools the crew may call, how much it may spend, and what its output must satisfy before it ships. On the connect path the crew's tools fire in your process, so the boundary is the spend caps, the output validators, and the record of what happened.