| home |
| name |
text |
tagline |
image |
actions |
| MateClaw |
The personal AI your IT department can actually sign off on. |
Other personal AI agents are built for one person. MateClaw is built for a team — multi-user workspaces, approval-gated sensitive actions, full audit trail, production-grade health monitoring. One self-hosted JAR; you control persisted data and outbound integration boundaries. |
| src |
alt |
| /logo.png |
MateClaw |
|
| theme |
text |
link |
| brand |
Get Started → |
/en/quickstart |
|
| theme |
text |
link |
| alt |
Read the Docs |
/en/intro |
|
|
|
| theme |
text |
link |
| alt |
3.0.0 Snapshot Changelog |
/en/releases/3.0.0 |
|
|
|
| icon |
title |
details |
| ⚙️ |
3.0.0-SNAPSHOT — Execution architecture upgrade |
3.0.0-SNAPSHOT upgrades to Java 21, Spring Boot 4 and Spring AI 2. AgentScope 2 drives ReAct and plan-step execution loops; MateClaw retains approvals, permissions, Goals, SSE and persistence. This is a development snapshot. |
|
| icon |
title |
details |
| ✅ |
2.3.0 — Managed JSON acceptance |
Users select required top-level fields; the server binds checks to current requirements and immutable JSON versions. Execution evidence defaults to observation, and ordinary artifact diagnostics do not replace acceptance bindings. |
|
| icon |
title |
details |
| ⚙️ |
Employee runtimes, not one Agent loop |
2.2.0 uses a shared Runtime Contract to decouple employee identity from execution. Native and DeepSeek Harness share conversations, workspaces, tool governance, and event projection; durable Goals recover across requests and backend restarts, while A2A connects external agents. |
|
| icon |
title |
details |
| 🧑💼 |
Digital employees, not chatbots |
You hire coworkers, not a chat box. Each one has a role, a goal, a backstory, a pixel-art avatar, and a color of their own — six built-in templates ship ready to use. ReAct + Plan-and-Execute, parallel delegation between employees. |
|
| icon |
title |
details |
| 🤝 |
Teams, not lone wolves |
One team request becomes one Team Run — objective, task DAG, worker execution, final synthesis, and deliverables stay together. Chat delivers, Agents observes, and Teams governs; 2.2.0 further hardens checkpoint recovery and deliverable completion gates. |
|
| icon |
title |
details |
| 🧩 |
Skills are the skeleton, not a plugin |
One SKILL.md plus one LESSONS.md that grows with use. Eight starter templates, a five-step creation wizard, pre-flight checks before install. MCP and ACP bridges — even Claude Code and Codex show up as employees. |
|
| icon |
title |
details |
| 📚 |
Knowledge, shaped |
The LLM Wiki digests raw files into structured pages with summaries and backlinks. A library you can read, not a vector store you query. The hot cache auto-injects into your employees' system prompts. |
|
| icon |
title |
details |
| 🧬 |
Memory that compounds |
Session context, post-chat extraction, workspace files, and scheduled Dreaming consolidation. Tomorrow's conversation starts where today's ended. |
|
| icon |
title |
details |
| 👀 |
You see what every employee is doing |
The Admin Runtime Console shows who's running, what step they're on, how many tokens they've used, with a one-click force-recycle when stuck. Streaming is staged honestly, multi-agent delegation no longer fights itself, long tasks demand evidence-grounded answers. |
|
| icon |
title |
details |
| 🔀 |
Business processes, not manual hand-offs |
Workflows compose multiple employees plus system actions (approvals, channel dispatch, write-memory) into a publishable, triggerable, replayable linear DSL — seven step modes. Triggers wire system events to those flows — six pattern types cover cron, webhooks, channel messages, employee lifecycle, content match, and workflow completion. |
|
| icon |
title |
details |
| 🌐 |
Every surface that matters |
Web console, desktop app with bundled JRE 21, and eight chat channels. Same brain, same memory, wherever your team works. |
|
|
