awesome-openclaw-agents and opencrew

These are **complements**: one provides reusable agent configuration templates (SOUL.md configs) while the other provides the multi-agent orchestration framework and Slack integration to deploy and coordinate those agents in production.

awesome-openclaw-agents
52
Established
opencrew
50
Established
Maintenance 13/25
Adoption 10/25
Maturity 9/25
Community 20/25
Maintenance 13/25
Adoption 10/25
Maturity 9/25
Community 18/25
Stars: 331
Forks: 50
Downloads:
Commits (30d): 0
Language: JavaScript
License: MIT
Stars: 269
Forks: 39
Downloads:
Commits (30d): 0
Language: Shell
License: MIT
No Package No Dependents
No Package No Dependents

About awesome-openclaw-agents

mergisi/awesome-openclaw-agents

Curated list of AI agent templates for OpenClaw. Ready-to-use SOUL.md configs for productivity, development, marketing, and business agents. Deploy with CrewClaw.

# Technical Summary Provides 187 production-ready SOUL.md agent configurations across 24 categories (productivity, development, DevOps, finance, etc.) that define agent behavior, tools, and integrations within the OpenClaw framework. Each template is copy-paste ready with optional JSON machine-readable exports, and integrates with CrewClaw's deployment platform to generate complete Docker packages without manual setup. Supports MCP server connections and enterprise integrations, enabling rapid agent instantiation via `openclaw agents add` CLI or web-based configuration without terminal access.

About opencrew

AlexAnys/opencrew

Openclaw多智能体协同系统 | Multi-Agent OS for Decision Makers — 基于 OpenClaw (Clawbot) + Slack,让 AI 团队各司其职、自主稳定迭代。

Organizes multi-agent teams through chat platforms (Slack, Feishu, Discord) with a three-layer architecture: CoS (Chief of Staff) for intent alignment, domain specialists (CTO/Builder/CIO/Research) for execution, and KO/Ops for knowledge consolidation and drift prevention. Implements Agent-to-Agent (A2A) coordination through a two-step protocol with autonomy levels (L0-L3) determining when agents self-execute versus escalate, and structures task knowledge into three layers—raw conversations, closeout summaries (~25x compression), and reusable organizational patterns extracted by the KO agent.

Scores updated daily from GitHub, PyPI, and npm data. How scores work