Kapso

Kapso is a Python engine that runs experiments toward a goal, deploys the best solution, and keeps lessons from each run for machine learning developers.

What it is good for

  • Finding the best machine learning approach through automated experiments
  • Deploying a trained solution to Docker, Modal or similar targets
  • Reusing lessons from past experiments on new projects
  • Building predictive models on relational data

At a glance

Who it is for
Setup difficulty
Technical ยท Guided command-line setup

Level 4 of 5. You copy and paste commands into a terminal by following the instructions. Comfortable territory for vibe coders.

Cost
Free and open source
Type
Framework or SDK
Tasks
Works with
Claude Code, Codex, Docker, Modal, BentoCloud, LangGraph Platform

How to get started with Kapso

  1. Install Node.js 18 or newer, then install the Codex and Claude Code command-line tools and log in to both.
  2. Add a Leeroopedia key to your .env file as LEEROOPEDIA_API_KEY (recommended).
  3. Install the package with pip install leeroo-kapso (Python 3.10 or newer).
  4. Run kapso doctor to check the setup and see fixes for anything missing.
  5. Optionally install the plugin in Claude Code, run /kapso:setup, then /kapso:build <what you want>.

From the project's own documentation. If a step does not match what you see, the official pages have the latest instructions.

Before you try it

  • Needs Node.js plus both Codex and Claude Code command-line tools logged in; there is no direct API fallback.
  • An OpenAI key is optional and only used for knowledge graph and semantic search features.
  • Leeroopedia is a hosted service; a new account comes with $20 of free credit.
  • Requires Python 3.10 or newer.

Momentum

Rank this week
#18 in builder tools
Best rank
#18 week 46, 2026
Stars gained last week
+87
GitHub stars
226

GitHub numbers as of Nov 16, 2026. Updated weekly.

127 180 233 Oct 29, 2026 Nov 16, 2026
GitHub stars for Kapso: 134 on Oct 29, 2026, 226 on Nov 16, 2026. Source: GitHub.

Ranking history

First ranked in week 46, 2026.

Questions

Is Kapso free?
Yes. It is free and open source (MIT license).
Do I need to know how to code to use Kapso?
You do not need to write code, but setup means copying and pasting commands into a terminal by following its instructions.
What does Kapso work with?
According to its own documentation, it works with Claude Code, Codex, Docker, Modal, BentoCloud, and LangGraph Platform.
How do I get started with Kapso?
1. Install Node.js 18 or newer, then install the Codex and Claude Code command-line tools and log in to both. 2. Add a Leeroopedia key to your .env file as LEEROOPEDIA_API_KEY (recommended). 3. Install the package with pip install leeroo-kapso (Python 3.10 or newer). 4. Run kapso doctor to check the setup and see fixes for anything missing. 5. Optionally install the plugin in Claude Code, run /kapso:setup, then /kapso:build <what you want>.
What should I know before trying Kapso?
Needs Node.js plus both Codex and Claude Code command-line tools logged in; there is no direct API fallback. An OpenAI key is optional and only used for knowledge graph and semantic search features. Leeroopedia is a hosted service; a new account comes with $20 of free credit. Requires Python 3.10 or newer.
What is Kapso good for?
Finding the best machine learning approach through automated experiments, deploying a trained solution to Docker, Modal or similar targets, reusing lessons from past experiments on new projects, and building predictive models on relational data.

Similar agents to Kapso

  • RD-Agent

    Automates data-driven AI research and development, with less emphasis on deployment and a scored bank of lessons.

  • harness2

    Refines the prompts, tools and skills around an agent rather than running experiments and deploying solutions.

  • opal

    Lets a coding agent rewrite the optimizer during model training, a narrower scope than whole experiment campaigns.

  • Agent Zero

    General-purpose agent that works on its own Linux computer, not focused on machine learning experiments.

  • pydantic-ai

    Code library for building agents with validated outputs, with no experimentation or lesson learning built in.

  • Kestra

    Kestra is an open-source orchestration platform that schedules and automates data, infrastructure, and AI agent workflows for technical teams.

Also compared with Kapso on these pages:

Last updated Nov 16, 2026. Tracked since Oct 29, 2026. Parts of this summary required judgment because the source material was brief. Spotted an error? Tell us.