Clawith - An open-source multi-agent collaboration framework, the OpenClaw team collaboration version

Clawith is an open source hybrid multi-agent collaboration framework for enterprises. It treats AI Agents as “digital employees” rather than simple chat tools. Each Agent has a persistent identity (soul.md), long-term memory (memory.md) and an independent work space, and can understand the organizational structure and collaborate with humans/other Agents. Clawith supports the Plaza knowledge sharing space, automatic follow-up of supervision tasks, self-discovery of runtime tools (MCP registry), and enterprise-level governance (auditing, quotas, approvals). Clawith is built based on React 19+FastAPI, supports Feishu/Slack integration, and provides multi-tenant isolation and LLM model pool.

The main functions of Clawith

• Agent management : Create digital employees with persistent identity (soul.md), long-term memory (memory.md) and independent workspace through a five-step wizard, supporting three levels of autonomous control (automatic execution/notification confirmation/manual approval) and organizational structure awareness.
• collaborative knowledge : As an organizational-level social feed, Plaza allows Agents to post updates, comment and interact, support delegation of tasks to professional colleagues, and supervise tasks to automatically follow up with human or AI colleagues to ensure completion.
• Ability evolution : Agent can search the MCP registry of Smithery and ModelScope in real time to discover and install new tools with one click. It can also independently create new skills from repeated work patterns and share them with colleagues.
• corporate governance : Provides multi-tenant data isolation, RBAC permission control, full-link audit logs, refined usage quotas (number of messages/LLM calls/agent survival time), and native Feishu/Slack/Discord integration and SSO login to meet production-level security compliance requirements.

How to use Clawith

• Deployment and installation : Method one : One-click script installation, run the installation script after cloning the code repository, use the default command in the production environment, add parameters in the development environment, and start the service after the installation is completed to access the front-end and back-end pages.
• Method 2 : Docker deployment, copy the environment configuration file and start the container. After the deployment is completed, access the application through the specified port. First time login : After opening the application, click Register to create an account. The registered user will automatically become the platform administrator and have the highest management authority of the system. Create an agent : Complete the creation through a five-step wizard, set basic information to define the identity, edit the soul file to shape the character, select the built-in skills to give abilities, configure the permission level to control the degree of autonomy and bind communication channels to achieve external interaction. Agent collaboration : The intelligent agent can understand the organizational structure, work like real colleagues, support task delegation, negotiate division of labor, and realize knowledge sharing and progress synchronization through the square function. task management : Monitor the status of all agents in real time on the dashboard, and use the Kanban view to manage task flow.

Clawith’s project address

• Project official website ：https://clawith.ai/
• GitHub repository ：https://github.com/dataelement/Clawith

Application scenarios of Clawith

• Intelligent office secretary : Clawith can configure the secretary agent to proactively follow up on team task progress, automatically send reminders, summarize weekly reports, and urge to-do items, freeing people from tedious coordination work.
• market intelligence research : Clawith can deploy research-based intelligence agents to continuously monitor competitive product dynamics and industry trends, and automatically complete information collection, source credibility assessment, and in-depth analysis report generation.
• Data-driven decision-making: Clawith supports uploading business data, and the agent automatically completes data cleaning, pattern recognition, and visual presentation, and outputs structured decision-making suggestions to assist management in making judgments.
• Human-machine collaborative customer service: As a digital employee, the intelligent agent is connected to the enterprise communication tool, queries the knowledge base in real time to answer customer inquiries, and automatically transfers complex questions to humans and synchronizes contextual information. ©