Runsight is an open-source, self-hosted YAML-first workflow engine for AI agents that lets you design, version and run agent pipelines as Git-native YAML files or on a visual canvas; it orchestrates LLM/API calls into reusable blocks with per-step cost and latency tracking, hard budget caps, pause/kill controls, and real-time state inspection. It adds AI-specific features like assertions, transform hooks and regression testing to evaluate and debug agent outputs systematically, while keeping your API keys and models on-premise. Use it to turn opaque, ad-hoc agent code into reproducible, reviewable, cost-controlled AI workflows that integrate with standard Git development practices.
Design agent workflows in YAML. Commit to Git. Track cost per run. Evaluate with built-in assertions. Pause or kill agents mid-flight.
Comments will not be approved to be posted if they are SPAM, abusive, off-topic, use profanity, contain a personal attack, or promote hate of any kind.