AgentToolkit/altk-evolve

README preview

<div align="center">

# Evolve: On‑the‑job learning for AI agents

[![Python](https://img.shields.io/badge/python-3.12%2B-blue)](https://www.python.org/)
![Status](https://img.shields.io/badge/status-active-brightgreen)
[![Documentation](https://shields.io/badge/Official%20Webpage-Documentation-blue)](https://agenttoolkit.github.io/altk-evolve)
[![arXiv](https://img.shields.io/badge/arXiv-2603.10600-b31b1b)](https://arxiv.org/pdf/2603.10600)
[![License](https://img.shields.io/github/license/AgentToolkit/altk-evolve)](https://www.apache.org/licenses/LICENSE-2.0)
![Stars](https://img.shields.io/github/stars/AgentToolkit/altk-evolve?style=social)

**Blog posts:** [IBM announcement](https://www.ibm.com/new/announcements/altk-evolve-on-the-job-learning-for-ai-agents) | [Hugging Face blog](https://huggingface.co/blog/ibm-research/altk-evolve)
</div>

Coding agents repeat the same mistakes because they start fresh every session. Evolve gives agents memory — they learn from what worked and what didn't, so each session is better than the last.

Evolve is a system designed to help agents improve over time by learning from their trajectories. The Lite version is designed to effortlessly slot into existing agent assistants like Claude Code and Codex. It uses a combination of an MCP server for tool integration, vector storage for memory, and LLM-based conflict resolution to refine its knowledge base.

On the AppWorld benchmark, Evolve improved agent reliability by +8.9 points overall, with a 74% relative increase on hard multi-step tasks. Evolve is a system designed to help agents improve over time by learning from their trajectories. It uses a combination of an MCP server for tool integration, vector storage for memory, and LLM-based conflict resolution to refine its knowledge base.

View full repository on GitHub →