Belay

Durable execution for AI agents in Rails.

Sidekiq gives you durable background jobs. Belay gives you durable AI jobs — every model call and tool call is a database row, so a run can crash at step 47 and resume at step 47.

It sits under whatever prompting style you use and records everything. It is not a prompt library, not a chain-builder DSL, and not a wrapper that hides your LLM SDK.

Status: early release (v0.1.0). Core runtime, gates & approvals, replay, caching & rules, and the dashboard are implemented and proven with real process kills and real concurrent workers, not just designed - see the design decisions.

Where to start

  • Gates and approvals - pausing a run for a human, approve!/reject!, and needs_review’s crash-recovery pause.
  • Rules - how a consistently-approved gate learns to stop pausing.
  • Replay - re-running a completed run’s model calls against a different model.
  • Caching - cacheable: true and when it’s actually safe to use.
  • Providers - Belay isn’t Anthropic-only; bring your own.
  • Dashboard - mounting the engine and wiring in your own auth.

For installation, configuration, and the full workflow DSL, see the README.


This site uses Just the Docs, a documentation theme for Jekyll.