# Memento-Teams/Memento-Skills

Memento-Skills: Let Agents Design Agents

Repository: https://github.com/Memento-Teams/Memento-Skills
Canonical: https://ross.abutalabs.com/products/memento-skills
Homepage: https://arxiv.org/abs/2603.18743
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-17T03:41:43+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 86, longevity 14
- inputs: {"age_days": 201, "days_push": 16, "days_rel": 16, "gap_med": 58.0, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1546, forks 180 (observed 2026-08-28T04:05:01.587548+00:00)

## What it is
Memento-Skills is a generalist, continually-learnable LLM agent framework where reusable skills stored as markdown files serve as persistent, evolving memory. It acts as an agent-designing agent, autonomously constructing and improving task-specific agents through a Read-Write Reflective Learning loop without updating LLM parameters.

## Use cases
- build an agent that learns skills from experience
- let an AI agent design other agents for new tasks
- run a chat agent with persistent reusable skills
- continual learning for LLM agents without fine-tuning
- deploy an agent with web research and terminal skills
- connect an agent to Feishu, DingTalk, WeCom, or WeChat
- run multi-step tasks with reflection-based replanning

## When to choose
- you want an agent framework with persistent, extensible skill libraries
- you need continual learning without retraining model weights
- you want an agent that can compose or design other agents
- you need CLI plus desktop app access with IM platform integrations

## When to avoid
- you need a simple one-shot chatbot without memory or skill evolution
- you require a mature production framework with a long track record
- you need non-Python runtimes or mobile/embedded deployment
- you want full control over low-level agent orchestration primitives

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, machine-learning, cli, gui
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: python, windows, cli, cross-platform
- tags: skill-centric-runtime, continual-learning, stateful-prompts, agent-designing-agent, memory-based-learning, reflection, desktop-app, im-integrations, ai-agents, macos, desktop

## Member repositories
- Memento-Teams/Memento-Skills (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:01.587548+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:30:32.760933+00:00, confidence not recorded.
  - readme: https://github.com/Memento-Teams/Memento-Skills (fetched 2026-08-28T04:05:01.587548+00:00, sha c6e2880c8878)
  - homepage: https://arxiv.org/abs/2603.18743 (fetched 2026-08-29T11:31:51.826578+00:00, sha cb2f80491eb6)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T11:31:51.835565+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T11:31:51.838795+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T11:31:51.840460+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T11:31:51.837195+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.