# Runsight

> Runsight is an open source, Git-native workflow engine for AI agents. It helps engineers design workflows in YAML, commit them to Git, track cost per run, and evaluate behavior with built-in assertions.

Runsight is for engineers who want readable workflow files, local-first control, and a practical way to design, run, and observe AI workflows.

Start here:
- [Homepage](https://runsight.ai/): product overview, positioning, and high-level capabilities.
- [Documentation](https://runsight.ai/docs/): complete product docs.
- [Quickstart](https://runsight.ai/docs/getting-started/quickstart/): fastest path to install and run Runsight.

Use the documentation sets below when an implementation task needs more detail than the homepage summary.

## Documentation Sets

- [Abridged documentation](https://runsight.ai/llms-small.txt): a compact version of the documentation for Runsight, with non-essential content removed
- [Complete documentation](https://runsight.ai/llms-full.txt): the full documentation for Runsight

## Notes

- The complete documentation includes all content from the official documentation
- The content is automatically generated from the same source as the official documentation

## Optional

- [Homepage](https://runsight.ai/): Product overview and positioning
- [Documentation](https://runsight.ai/docs/): Complete product documentation
- [Quickstart](https://runsight.ai/docs/getting-started/quickstart/): Fastest path to install and run Runsight
- [AI integration guide](https://runsight.ai/skills.md): Machine-readable API and integration context for coding assistants
- [GitHub](https://github.com/runsight-ai/runsight): Source code, issues, and contribution context