Open-source multi-agent LLM framework that simulates a trading firm, with analyst, researcher, trader and risk agents debating market decisions.
dataelement/Clawith
Open-source multi-agent collaboration platform where each AI agent has a persistent identity, long-term memory, and its own workspace, working as a team.
- Stars
- 4,174
- Forks
- 697
- Open issues
- 259
- Last commit
- 2 months ago
What is it?
Clawith is a multi-agent collaboration platform aimed at organizations. Each agent keeps a persistent identity defined by a soul.md personality file, a memory.md long-term memory file, and a private file system with sandboxed code execution, so it stays consistent across conversations. Agents appear in an org chart, can message each other, delegate tasks, and post updates to a shared feed called the Plaza.
Agents manage their own schedules through an Aware system. They can create, adjust, and remove triggers of six types: cron, once, interval, poll (HTTP monitoring), on_message, and webhook. Each task-related trigger is bound to a focus item, so triggers are cancelled when the work is finished.
The platform includes multi-tenant RBAC, per-agent Slack, Discord, or Feishu/Lark bot identities, usage quotas, approval workflows for dangerous operations, audit logs, and a shared knowledge base. Agents can also discover and install tools at runtime through Smithery and ModelScope. It runs on a React 19 frontend and a FastAPI backend, with SQLite or PostgreSQL and Docker support.
Key features
- 01Persistent agent identity via soul.md and memory.md files
- 02Six trigger types, including cron, webhook, and HTTP polling
- 03Org-chart-aware agents that message and delegate to each other
- 04Per-agent Slack, Discord, and Feishu/Lark bot identities
- 05Multi-tenant RBAC, usage quotas, approval workflows, and audit logs
- 06Runtime tool discovery via Smithery and ModelScope
Use cases
- Running a team of AI agents that each own recurring monitoring or reporting tasks
- Deploying agents inside a company that answer through Slack, Discord, or Feishu/Lark
- Building an internal AI workforce with approval gates for risky actions
Quick start
- 1
Clone the repository
git clone https://github.com/dataelement/Clawith.git - 2
Enter the project directory and run the one-command setup
cd Clawith && bash setup.sh - 3
Start the app, then open http://localhost:3008
bash restart.sh
Things to know
- Requires external LLM API keys from providers such as OpenAI or Anthropic, since no models run locally.
- Minimum setup needs Python 3.12+, Node.js 20+, and a PostgreSQL or SQLite database.
In collections
Topics
FAQ
- What is Clawith?
- Open-source multi-agent collaboration platform where each AI agent has a persistent identity, long-term memory, and its own workspace, working as a team. Clawith is a multi-agent collaboration platform aimed at organizations. Each agent keeps a persistent identity defined by a soul.md personality file, a memory.md long-term memory file, and a private file system with sandboxed code execution, so it stays consistent across conversations. Agents appear in an org chart, can message each other, delegate tasks, and post updates to a shared feed called the Plaza.
- Is Clawith free to use? What license does it have?
- Clawith is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is Clawith still maintained?
- Updated within 6 months. The last commit to Clawith was on Aug 27, 2026.
- What can I use Clawith for?
- Running a team of AI agents that each own recurring monitoring or reporting tasks Deploying agents inside a company that answer through Slack, Discord, or Feishu/Lark Building an internal AI workforce with approval gates for risky actions
- How do I install Clawith?
- Clone the repository: git clone https://github.com/dataelement/Clawith.git Enter the project directory and run the one-command setup: cd Clawith && bash setup.sh Start the app, then open http://localhost:3008: bash restart.sh
- Where can I get Clawith?
- The source code and installation instructions are on GitHub: https://github.com/dataelement/Clawith
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Self-hosted AI operating system that runs a workforce of AI agents on your own machine, with any model, 1,000+ tool integrations and a Knowledge Graph.
Self-hosted AI stock monitor for A-shares, Hong Kong and US stocks, with portfolio tracking, price alerts, TradingAgents-based analysis and automated reports.
An AI-assisted novel writing app that helps authors generate outlines, characters, worldbuilding and chapters, self-hosted with FastAPI, React and PostgreSQL.
Open-source multi-agent platform that gives each AI agent its own cloud workspace with filesystem, desktop, browser, network and long-term memory.
An open agent framework where the AI agent gets its own Dockerized Linux desktop, browser, and document tools to carry out real work.