MateClaw is a self-hosted team AI operating system designed to keep organizational AI processes and data within the user's own infrastructure. It offers a unified platform that brings together autonomous AI agents, a self-maintaining wiki knowledge base, and memory management with per-user isolation, all running from a single deployable JAR file. The tool emphasizes verifiable and recoverable agent actions, automated wiki graph maintenance, and memory boundaries that separate individual user data. Its architecture is structured in four layers, covering goal execution, knowledge management, personalized memory, and runtime governance.
Key features include agent orchestration with criteria-based checklists for task completion, automated replanning and delegation, and a visual Kanban board for tracking multi-step plans. The wiki component supports automated entity extraction, customizable entity types, permission controls, and a knowledge graph that self-repairs broken links and propagates stale status. Memory is managed with owner-based isolation, visibility controls (personal, team, global), and safeguards such as injection budgets, nightly consolidation, and file limits to prevent uncontrolled context growth.
MateClaw integrates with various communication channels, including web, desktop, enterprise IM platforms (such as Feishu, DingTalk, WeChat Work, WeChat, Telegram, Discord, Slack), and embeddable web chat widgets. It supports desktop applications for Windows, macOS (Apple Silicon recommended), and Linux (AppImage), as well as deployment via Docker, JAR, or source code. The system is built on technologies like Spring Boot, Vue 3, TypeScript, MySQL, and Electron, and is open-source and auditable. Administrators have access to a centralized console for system health, agent management, tool governance, and channel configuration.
Additional capabilities include a "Content Studio" workflow for producing and publishing content to platforms like WeChat Official Accounts and Xiaohongshu, with features for AI de-biasing, compliance scanning, and automated formatting. The platform also provides skills management, external tool integration via MCP and ACP protocols, and runtime approval workflows for high-risk actions. MateClaw is positioned as a comprehensive, self-contained AI operating system for organizations that require private, auditable, and controllable AI collaboration and knowledge management.
MateClaw sits in PulseGate's AI & ML category. Enabling teams to run private, verifiable AI agents and knowledge management on their own infrastructure. It is built as an open-source project for technical teams and organizations needing private AI collaboration. MateClaw is open source under the Apache-2.0 license. The product ships for the web, the command line, macOS, Windows, and Linux, and it can be self-hosted.
It is developed by MateClaw Team, and the product first shipped in 2026. Development happens publicly on GitHub with 666 stars and 1.2k commits in the last 90 days. PulseGate's similarity index finds few close equivalents — MateClaw occupies a relatively distinct niche. Key capabilities include Self-hosted AI OS, team memory isolation, and knowledge graph wiki. It exposes integrations via an MCP server.
Latest indexed changes and source events
Other apps tracked under the same category.
