Open Source
Kapso
Kapso by Leeroo: Long-running agents that optimize AI and Data systems, and learn from every experience.
Our verdict
We have not reviewed Kapso yet
No one at Enterprise DNA has checked this listing against its sources. The description comes from public listings gathered automatically. Any fact below shows where it came from and when it was last checked.
From the listing
About Kapso
Kapso is an open-source Python project from Leeroo that provides long-running agents for optimizing AI and data systems, with a focus on learning from each experience. It is ranked #1 open-source on MLE-Bench, ALE-Bench, and RelBench benchmarks.
Best for
Teams building long-running AI/data optimization agents that want a benchmark-proven open-source starting point.
Use cases
- Automate optimization of machine learning pipelines
- Monitor and improve data system performance over time
- Benchmark agent performance on MLE-Bench, ALE-Bench, and RelBench
Notes
118 stars on GitHub. Last updated 2026-09-29. Licensed MIT.
Pros
- Open-source with a Python codebase
- Proven performance on multiple benchmarks
- Designed for long-running autonomous operation
Cons
- Small user base (118 stars)
- Focused on benchmark performance, not general-purpose observability
- Limited information available beyond benchmark claims
Key facts
What we have checked
- Days since the last commit
- 0
- GitHub forks
- 21
- GitHub stars
- 138
- Days to a first reply on issues
- 17.5
- Open issues
- 5
- Releases in the last 90 days
- 41
Each value shows when we last confirmed it and where it came from. Values we have not confirmed are left out, never guessed.
Stars and usage
How Kapso is growing
118 to 138
Health and security
Is Kapso looked after, and what do its setup files show
Health score
87 out of 100
| Part | Points | What we saw |
|---|---|---|
| Recent activity | 30 of 30 | Last push 0 days ago |
| Releases | 15 of 15 | 41 releases in 90 days |
| Issue replies | 9.3 of 20 | First reply in 17.5 days (median) |
| Maintainer | 15 of 15 | Owned by an organisation |
| Official source | Not counted | MCP servers only |
Sources
Where this comes from
- GitHubRegistry
A source shows what was published when we checked it. It does not certify security, performance or a working installation. Read our method.
Changelog
What changed
No changes recorded yet. Releases, price changes and deprecations appear here, with their dates, once we track them for this record.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.