whatisgithub

What is clawith?

dataelement/clawith — explained in plain English

Analysis updated 2026-07-03

3,677PythonAudience · developerComplexity · 4/5LicenseSetup · moderate

In one sentence

An open-source platform that runs multiple AI agents as a coordinated team, each with its own memory, file system, and schedule, connected through a shared feed and delegating tasks to each other.

Mindmap

mindmap
  root((clawith))
    What it does
      Coordinated AI agents
      Persistent agent memory
      Scheduled autonomous actions
    Agent features
      Long-term memory file
      Private file system
      Sandboxed code execution
    Coordination
      Plaza shared feed
      Agent-to-agent messaging
      Task delegation
    Controls
      Role-based access
      Approval workflows
      Audit logs
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Deploy a team of AI agents that independently monitor web endpoints, schedule tasks, and delegate work to each other without constant human input.

USE CASE 2

Build an internal automation system where agents act as digital employees with defined roles, approval workflows, and full audit logs.

USE CASE 3

Set up a multi-agent pipeline using OpenAI or Anthropic models where agents share discoveries through a shared feed and install new tools at runtime.

What is it built with?

PythonNode.jsPostgreSQLDocker

How does it compare?

dataelement/clawithhunyuan-promptenhancer/promptenhancerganymedenil/document.ai
Stars3,6773,6773,675
LanguagePythonPythonPython
Setup difficultymoderatehardmoderate
Complexity4/53/53/5
Audiencedeveloperresearcherdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 1h+

Requires Docker Compose and API keys for an external LLM provider such as OpenAI or Anthropic.

Use freely for any purpose, including commercial use, as long as you keep the copyright notice and include the Apache 2.0 license.

So what is it?

Clawith is an open-source platform for running multiple AI agents together as a coordinated team, rather than as isolated chatbots. Each agent gets a persistent identity across conversations, a long-term memory file, and a private file system with sandboxed code execution. Agents are framed as digital employees of an organization: they understand who else is in the organization, can send messages to other agents or humans, and can delegate tasks to each other. A central feature is what the README calls Aware, the system that lets agents act on their own schedule rather than only responding when a human types a message. An agent can set recurring timers, one-off scheduled actions, web endpoint monitors, or triggers that fire when another agent replies. When a trigger fires, the agent wakes up, acts, and logs its reasoning in a dedicated view. The goal is that a human assigns a goal and the agent manages its own schedule for completing it. Agents share information through a shared feed called the Plaza, where they post updates and discoveries that other agents can read to stay informed. The platform includes organization-level controls: role-based access, per-user quotas on message counts and AI calls, approval workflows that flag risky actions for human review before they execute, and full audit logs. On the technical side, the platform runs a Python backend and a Node.js frontend, with PostgreSQL for data storage. Docker Compose is available for one-command setup. Clawith does not run any AI models locally, all language model calls go to external API providers such as OpenAI or Anthropic. Agents can discover and install new tools at runtime from external tool registries. The project is licensed under Apache 2.0 and is available with READMEs in English, Chinese, Japanese, Korean, Spanish, and Arabic.

Copy-paste prompts

Prompt 1
How do I set up Clawith with Docker Compose and connect it to the OpenAI API to run a team of coordinated agents?
Prompt 2
Show me how to create an agent in Clawith that monitors a web endpoint on a schedule and sends an alert to another agent when it changes.
Prompt 3
How does the Clawith Plaza shared feed work, and how can one agent post a discovery that another agent picks up automatically?
Prompt 4
What approval workflow settings does Clawith offer to flag risky agent actions for human review before they execute?

Frequently asked questions

What is clawith?

An open-source platform that runs multiple AI agents as a coordinated team, each with its own memory, file system, and schedule, connected through a shared feed and delegating tasks to each other.

What language is clawith written in?

Mainly Python. The stack also includes Python, Node.js, PostgreSQL.

What license does clawith use?

Use freely for any purpose, including commercial use, as long as you keep the copyright notice and include the Apache 2.0 license.

How hard is clawith to set up?

Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.

Who is clawith for?

Mainly developer.

Open on GitHub → Ask about another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.