Learn from failure
Failures are treated as training signals, not just reasons to retry.
What Is Memento-Skills?
Memento-Skills is a self-evolving agent framework built around skills as first-class
units of capability. Instead of treating tools as a flat list, it treats them as a library the agent can retrieve,
execute, evaluate, repair, and rewrite over time.
The core loop is simple: Read - Execute - Reflect - Write. When a task fails, the system does not just retry. It tries to identify the weak skill, improve it, and write the improved capability back into memory.
The three paradigms of LLM adaptation. Pre-training and fine-tuning update model parameters and demand large data and compute budgets. Deployment-time learning keeps the model frozen while accumulating experience in external skill memory, enabling continual adaptation from live interactions.
The architecture of the self-evolving agent based on a Read - Execute - Reflect - Write loop. The system retrieves or generates skills, executes them, reflects on the outcome, and writes improvements back into the skill library.
Learning Results
We evaluate Memento-Skills on two challenging benchmarks: HLE and GAIA.
HLE
Humanity's Last Exam probes expert-level reasoning with extremely difficult questions across disciplines.
GAIA
GAIA evaluates general-purpose assistants on real-world, multi-step tasks involving tools, files, and the web.
Performance improves over multiple learning rounds while the skill library grows from atomic tools into a richer, more repairable and more retrievable memory of capabilities.
Core question. Memento-Skills is not centred on how to make an assistant run. It is centred on how to make an agent learn from deployment experience, reflect on failure, and rewrite its own skill code and prompts.
Rewrite its own skills
The system can optimise prompts, modify skill code, and create new skills when needed.
Run in the real world
Local execution, persistent state, CLI, GUI, and IM integration make it deployable beyond a paper demo.
Why It Matters
Not just tool use
The system is designed to improve weak skills instead of only accumulating more tools.
Built for deployment-time learning
It learns from live tasks while keeping the underlying model frozen.
Practical surfaces
CLI, GUI, Feishu integration, and local sandbox execution make it usable beyond a paper demo.
Deployment Surfaces
CLI, desktop GUI, Feishu bridge, skill verification, and local sandbox execution make the system practical for real deployment.
CLI and desktop GUI
memento agent and memento-gui cover both terminal-first and visual workflows.
Feishu and verification flow
memento feishu and memento verify extend the system into messaging and skill auditing.
One Repo. One Learning Agent.
python -m venv .venv && source .venv/bin/activate && pip install -e . && memento doctor && memento agentOne-Click GUI Install
Download the pre-built desktop app. No Python or terminal needed. Just unzip and run.
macOS (Apple Silicon)
Windows (x64)
Quick Start (Developer)
git clone https://github.com/Memento-Teams/Memento-Skills.git
cd Memento-Skills
python -m venv .venv
source .venv/bin/activate
pip install -e .On first launch, ~/memento_s/config.json is created automatically. Fill in your model profile, then start the app:
memento doctor
memento agent
memento-guiCitation
If you find Memento-Skills useful in your research, please cite:
@article{zhou2026mementoskills,
title={Memento-Skills: Let Agents Design Agents},
author={Zhou, Huichi and Guo, Siyuan and Liu, Anjie and Yu, Zhongwei and Gong, Ziqin and Zhao, Bowen and Chen, Zhixun and Zhang, Menglong and Chen, Yihang and Li, Jinsong and Yang, Runyu and Liu, Qiangbin and Yu, Xinlei and Zhou, Jianmin and Wang, Na and Sun, Chunyang and Wang, Jun},
journal={arXiv preprint arXiv:2603.18743},
year={2026},
url={https://arxiv.org/abs/2603.18743}
}