# modelscope/ms-agent

MS-Agent: a lightweight framework to empower agentic execution of complex tasks

Repository: https://github.com/modelscope/ms-agent
Canonical: https://ross.abutalabs.com/products/ms-agent
Homepage: https://ms-agent-en.readthedocs.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deep-research, agentic-insight, agentic-search, chat-bot, code-generation, memory
Last push: 2026-08-25T14:25:54+00:00

## Health v2 (maintenance only)
Score: 87/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 76, longevity 80
- inputs: {"age_days": 1126, "days_push": 8, "days_rel": 163, "gap_med": 13, "n_releases_24m": 18}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4369, forks 517 (observed 2026-08-28T04:08:46.713752+00:00)

## What it is
MS-Agent is a lightweight Python framework from ModelScope for building LLM-powered agents with autonomous exploration capabilities. It supports MCP-based tool calling, deep research workflows, code generation, agent skills, and a local WebUI with SSE streaming.

## Use cases
- build a chatbot that calls tools via MCP
- run deep research tasks with autonomous exploration
- generate code artifacts with an LLM agent
- create a multi-agent system for complex task execution
- add reusable skills and memory to an LLM agent
- generate short videos with an agent pipeline

## When to choose
- you want a lightweight, extensible Python agent framework with MCP support
- you need deep research or agentic search capabilities out of the box
- you want a local WebUI for interacting with agents
- you are in the ModelScope ecosystem and want integration with it

## When to avoid
- you need a heavyweight production multi-agent orchestration platform with enterprise features
- you are not working in Python or with LLMs
- you need a no-code agent builder for non-developers

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, chatbot, mcp, rag
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: python, cross-platform
- tags: deep-research, multi-agent, tool-calling, mcp-support, agentic-search, memory, short-video-generation, webui, code-generation, ai-agents, web-server

## Member repositories
- modelscope/ms-agent (main) score 87

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:46.713752+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 2026-08-29T18:21:23.429137+00:00, confidence not recorded.
  - readme: https://github.com/modelscope/ms-agent (fetched 2026-08-28T04:08:46.713752+00:00, sha 8f4ca67d1b8e)
  - registry_pypi: https://pypi.org/pypi/ms-agent/json (fetched 2026-08-29T09:10:30.109172+00:00, sha 4e40a8819f7a)
- Data as of 2026-08-30T08:39:29.467469+00:00.