#4141st of 102 in Agent platforms

clawith

Multi-agent platform with persistent agent identity, memory and org-chart awareness

Stars
4.2k
License
Apache-2.0
Last commit
Aug 2026
Last release
Aug 2026
Language
Python
Stars, last 29 weeks+882
25 Apr 2026: 3,309 stars2 May 2026: 3,425 stars9 May 2026: 3,542 stars16 May 2026: 3,667 stars23 May 2026: 3,767 stars30 May 2026: 3,866 stars6 Jun 2026: 3,888 stars13 Jun 2026: 3,925 stars20 Jun 2026: 3,959 stars27 Jun 2026: 3,976 stars4 Jul 2026: 3,994 stars11 Jul 2026: 4,013 stars18 Jul 2026: 4,036 stars25 Jul 2026: 4,049 stars1 Aug 2026: 4,064 stars8 Aug 2026: 4,076 stars15 Aug 2026: 4,089 stars22 Aug 2026: 4,098 stars29 Aug 2026: 4,109 stars5 Sept 2026: 4,115 stars12 Sept 2026: 4,124 stars19 Sept 2026: 4,133 stars26 Sept 2026: 4,142 stars3 Oct 2026: 4,151 stars10 Oct 2026: 4,155 stars17 Oct 2026: 4,158 stars24 Oct 2026: 4,173 stars31 Oct 2026: 4,178 stars7 Nov 2026: 4,186 stars14 Nov 2026: 4,191 stars

3,309 on 25 Apr 20264,191 on 14 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

Clawith is a FastAPI and React web app where each agent keeps a soul.md personality file, a memory.md file and a private sandboxed workspace. Agents run on six trigger types (cron, once, interval, poll, on_message, webhook), message each other and humans, and post to a shared feed. It adds multi-tenant RBAC, usage quotas, approval flows and Slack, Discord or Feishu/Lark channels. All LLM inference goes to external APIs.

Who it is for: Teams self-hosting a crew of persistent AI agents with org-level access control

Strengths

  • Six trigger types, including webhook and HTTP poll, that agents can create and cancel themselves
  • Multi-tenant RBAC, usage quotas, approval workflows and audit logs built in
  • Per-agent Slack, Discord or Feishu/Lark bot identity
  • Agents can install MCP tools from Smithery and ModelScope at runtime

Weaknesses

  • No local model inference; requires external LLM API endpoints
  • Needs PostgreSQL (SQLite for trials) and Redis; heavier setup than a single binary
  • Public demo is a shared environment, stability not guaranteed
  • Runtime tool installation by agents widens the attack surface; Docker socket access must be restricted

What it needs

  • RAM ≥ 4 GB
  • no GPU
  • Compose
  • Needs PostgreSQL 15+ or SQLite, Redis, Docker, External LLM API (OpenAI, Anthropic, etc.), Python 3.12+, Node.js 20+
  • Models: OpenAI, Anthropic, External LLM APIs
  • port 3008

What changed

  1. “Database migrations run automatically during application startup. No manual Alembic command is required. Back up the database and persistent files before upgrading production.”

  2. “Standard steps (no manual database commands required; migrations run automatically on startup).”

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Known problems

Reported, then fixed by a linked change

Titles are the reporters' own words, taken from the project's GitHub issues and ranked by reactions. We have not tried to reproduce any of them.

Also in Agent platforms

See all 102
Also in Agent platforms
RankProjectScore
1AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder Live demo ↗ (opens in a new tab)188.1k starscustom license95 out of 100
2daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps73.1k starsMIT95 out of 100
3LangflowVisual builder for AI agents and workflows, served as API or MCP server156.7k starsMIT94 out of 100
#4deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills85.5k starsMIT93 out of 100
#5DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs Live demo ↗ (opens in a new tab)161.4k starscustom license92 out of 100