# mateaix/mateclaw

🤖 MateClaw — Your second brain with Multi-Agent Orchestration, MCP Protocol, Skills & Memory, Dream, and Multi-Channel Support. Built on Spring AI Alibaba.

Repository: https://github.com/mateaix/mateclaw
Canonical: https://ross.abutalabs.com/products/mateclaw
Homepage: https://claw.mate.vip
Language: Java
License: Apache-2.0
License Family: permissive
Topics: ai-agent, dingtalk-robot, feishu-bot, mcp-protocol, multi-agent, openclaw, plan-and-execute, spring-ai, spring-boot, telegrambot, tool-calling, skills, agent, llm-wiki, multiagent, agent-harness, loop-engineering
Last push: 2026-09-02T09:56:21+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 99, longevity 10
- inputs: {"age_days": 151, "days_push": 0, "days_rel": 4, "gap_med": 9.0, "n_releases_24m": 17}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1072, forks 322 (observed 2026-09-03T02:15:14.758448+00:00)

## What it is
MateClaw is a self-hosted, team-oriented AI agent platform ('second brain') built on Spring Boot and Spring AI Alibaba, shipped as a single JAR or Docker deployment. It combines multi-agent orchestration (ReAct + Plan-and-Execute on a StateGraph runtime), MCP/ACP tool integration, skills with self-evolution, per-user memory, a wiki with knowledge graphs, and multi-channel messaging (Feishu, DingTalk, WeChat, Telegram, Discord, Slack, WebChat) with enterprise controls like approval gates, audit trails, and multi-vendor model failover.

## Use cases
- self-host a team AI assistant with multi-user workspaces
- run multi-agent plan-and-execute workflows with tool calling
- connect an AI agent to DingTalk, Feishu, WeChat, or Telegram
- build a team knowledge base wiki with citations and knowledge graphs
- give AI agents persistent per-user memory
- gate sensitive tool calls behind human approval with audit logs
- route LLM requests across multiple providers with failover
- automate content creation for WeChat official accounts and Xiaohongshu

## When to choose
- you need a deployable, auditable AI agent platform your IT/security team can approve
- you want multi-user team workspaces rather than a single-user personal agent
- your team communicates via Chinese enterprise IM platforms like Feishu, DingTalk, or WeChat
- you want a Java/Spring Boot stack with one-JAR deployment and full data control
- you need skills, memory, wiki, and MCP tools unified in one registry with governance

## When to avoid
- you want a lightweight personal agent with minimal setup
- you need a Python/TypeScript agent ecosystem rather than Java
- you only need a simple chatbot wrapper around one LLM API
- you require fully managed cloud hosting rather than self-hosting

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, chatbot, rag, mcp, chat-interface, webhook, self-hosted, workflow-automation, search-engine
- domain: large-language-models, chatbots, self-hosted, developer-tools, messaging-platforms
- platform: self-hosted, cross-platform, jvm
- tags: multi-agent-orchestration, plan-and-execute, spring-ai-alibaba, second-brain, team-workspaces, approval-gate, audit-trail, skills, memory, knowledge-graph, wiki, dingtalk, feishu, telegram-bot, discord, slack, wechat, model-failover, trajectory-replay, content-studio, ai-agents, automation, docker, web-server, desktop

## Member repositories
- mateaix/mateclaw (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:14.758448+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 2026-08-30T07:00:31.926322+00:00, confidence not recorded.
  - readme: https://github.com/mateaix/mateclaw (fetched 2026-09-03T02:15:14.758448+00:00, sha f9582a79add1)
  - homepage: https://claw.mate.vip (fetched 2026-08-29T13:02:01.237552+00:00, sha 1da4883ab1b3)
  - site_page: https://claw.mate.vip/docs (fetched 2026-08-29T13:02:01.240539+00:00, sha c8004106d160)
  - site_page: https://claw.mate.vip/docs/deepseek-harness (fetched 2026-08-29T13:02:01.242224+00:00, sha ad9139ad6a85)
  - site_page: https://claw.mate.vip/docs/zh/desktop (fetched 2026-08-29T13:02:01.243921+00:00, sha ad9139ad6a85)
  - site_page: https://claw.mate.vip/docs/zh/quickstart (fetched 2026-08-29T13:02:01.245686+00:00, sha ad9139ad6a85)
- Data as of 2026-08-30T08:39:29.467469+00:00.