Run LangGraph.js graphs on a governed runtime — Kohala
Import your TypeScript LangGraph.js graph onto Kohala's governed runtime — under a recorded charter, a server-enforced allowlist on the Kohala tools it can call, and a billing cap, unscheduled unless you ask for a cadence — or wrap it from your own Node app and report each run, capped, validated, metered, and traced.
Explore Kohala
Frequently Asked Questions
- Can I run a LangGraph.js graph on Kohala?
- Yes, either way. Paste your TypeScript graph 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 — arriving unscheduled, so it runs when you trigger it until you add a cadence. Or keep it in your own Node app and report each run with @kohala/sdk, metered against the same caps and checked by the same output validators. Both paths record credit metering and a trace ID.
- Is TypeScript a second-class citizen on the hosted runtime?
- No. The Node lane runs through the same admission gate, tool RPC, caps, validators, and audit trail as the Python one. Your declared npm dependencies are installed into an account-scoped bundle at deploy, and tokens your graph spends through its own provider key are metered like platform LLM calls.
- Does Kohala replace LangGraph.js state management?
- No. LangGraph.js keeps managing graph state and control flow. Kohala adds the layer around it: governance policies, cost metering, output validators, and an account-scoped shared metered memory API your graph can read and write across runs once you wire it in.