BYO computeNo token tax
03Pricing

Pay for the rails. Not the tokens.

RunGuard charges for orchestration, safety policy, and routing. Your LLM credits stay with you — Copilot, Claude, Cursor, or Codex.

04FAQ

Pricing questions

How does RunGuard route an incident?

When your app reports an error, RunGuard groups and classifies it, builds a crash packet with the stack trace and suspected files, and opens a GitHub issue in your repo. The issue is dispatched to the executor you picked — Copilot, Claude Code Action, Cursor, Codex, or a local agent — which uses your own LLM credits to propose a fix as a pull request.

Who pays for the LLM calls?

You do, through your existing Copilot seat, Anthropic API key, Cursor account, or OpenAI account. RunGuard charges only for orchestration, routing, and dashboard access — never for tokens or sandbox compute.

What counts as an error?

Each error occurrence counts once. If the same error happens 100 times, it counts as 100 errors. We deduplicate and group identical signatures so a single bug only creates one incident, not 100.

What if I need more than 1 repo on Starter?

Upgrade to Pro for 10 repositories, Team for 50, or Enterprise for unlimited.

Can I cancel anytime?

Yes. No contracts. You'll retain access through the end of your billing period.

Is there a free plan?

Yes. Starter is free with usage limits, so you can connect one repo and validate the workflow before choosing a paid plan.

Ready to stop firefighting?

Start free. Upgrade when your incidents earn their own line item.