# kensa

> Kensa is the open source harness for evaluating your agents.

## Docs

- [Changelog](https://kensa.sh/docs/changelog.md): What's new in kensa.
- [Checks](https://kensa.sh/docs/checks.md): Deterministic check types. Fast, cheap, binary.
- [CI Integration](https://kensa.sh/docs/ci.md): Run kensa in continuous integration pipelines.
- [CLI Reference](https://kensa.sh/docs/cli.md): All kensa commands and their options.
- [Concepts](https://kensa.sh/docs/concepts.md): The mental model behind kensa. Scenarios, spans, checks, judges, and results.
- [Example Agents](https://kensa.sh/docs/examples.md): Five example agents covering domains where reliability matters.
- [Introduction](https://kensa.sh/docs/introduction.md): Kensa is the open source harness for evaluating your agents.
- [Judge](https://kensa.sh/docs/judge.md): LLM-as-judge for natural-language criteria evaluation.
- [MCP Server](https://kensa.sh/docs/mcp-server.md): Serve kensa over the Model Context Protocol to any MCP-aware client.
- [Pytest Evals](https://kensa.sh/docs/pytest.md): Run Kensa scenarios through normal pytest tests and fixtures.
- [Quickstart](https://kensa.sh/docs/quickstart.md): Get kensa running in under a minute.
- [Scenarios](https://kensa.sh/docs/scenarios.md): YAML-based test scenarios that define what to run and how to judge it.
- [Skills](https://kensa.sh/docs/skills.md): Five skills that orchestrate the complete evals workflow.
- [Tracing & Instrumentation](https://kensa.sh/docs/tracing.md): OpenTelemetry-based tracing with auto-instrumentation for major LLM providers.

## OpenAPI Specs

- [openapi](https://kensa.sh/docs/api-reference/openapi.json)