An advanced multi-agent orchestration skill that automatically assembles specialized AI teams to handle complex development projects and technical tasks.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install claude-code-team
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install claude-code-team using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Claude Code Team?
Claude Code Team is a sophisticated coordination layer designed for developers who need more than a single agent response. By leveraging the Openclaw Skills framework, this tool automatically identifies project requirements and spawns a virtual team of specialized agents—including Product Managers, Architects, and Developers—each powered by the most cost-effective and capable models available like Qwen-Max and GLM-5.
The skill intelligently differentiates between casual queries and project-level tasks. When it detects complex triggers like bug fixes or system optimizations, it initiates a full-scale multi-agent workflow, ensuring that every aspect of the technical lifecycle is handled by a role-specific AI optimized for that exact function.
Claude Code Team Use Cases
- Automating full-stack software development cycles from requirement analysis to deployment.
- Managing complex bug squashing sessions that require code review and testing validation.
- Assembling specialized AI research squads or technical platform teams for architectural decision-making.
- Scaling productivity for solo developers by providing a 10-person virtual special ops team for rapid prototyping.
How Claude Code Team Works
- The system monitors user input for specific triggers such as software development team or project optimization.
- Upon activation, the skill identifies the required team configuration (Standard Dev, Technical Platform, or Custom).
- It automatically maps the most suitable LLM to each role (e.g., qwen-max for architects, glm-5 for programmers) to balance performance and cost.
- The skill spawns multiple concurrent sessions to execute role-specific responsibilities in parallel.
- All outputs are synthesized into a comprehensive final report and delivered to the user.
- Task history and lessons learned are persisted into a structured memory system for future context.
Claude Code Team Setup
To integrate this skill into your Openclaw Skills environment, ensure your model providers (Bailian, OpenAI, etc.) are configured.
# Initialize the team mode
openclaw-cli install claude-code-team
# Configure your model API keys
export BAILIAN_API_KEY='your_key_here'
Simply trigger the skill by mentioning a team name or asking to optimize a project.
Claude Code Team Data Schema & Taxonomy
The skill maintains a hierarchical memory structure to ensure continuity across sessions:
| Data Type | Storage Path | Description |
|---|---|---|
| Team Memory | memory/teams/{team_name}/MEMORY.md |
Stores task history, decision logs, and project-wide lessons. |
| Agent Memory | memory/agents/{role_name}/MEMORY.md |
Records individual agent growth, skill improvements, and task records. |
| Config Files | config/teams/ |
Stores JSON definitions for standard and custom team compositions. |
Claude Code Team Advanced Features
- Dynamic Model Routing: Automatically switches between qwen-max for complex reasoning and glm-5 for cost-efficient coding.
- Custom Team Construction: Ability to define bespoke teams like AI Research Squads with specific role responsibilities.
- Automated Persistence: Integrated memory system that tracks technical debt and architectural decisions across the project lifecycle.
- Performance Optimization: Built-in gateway timeout handling (120s) and concurrency management for large-scale agent deployments.
- Multi-Agent Synthesis: Sophisticated aggregation logic that combines outputs from up to 10 agents into a single actionable report.
SKILL.md
Loading
Related Openclaw Skills
Kiro Agentic IDE Workflow
Kiro is an Amazon-developed agentic IDE and CLI tool designed for spec-driven development using automated hooks, steering rules, and MCP integration.
Trip Guide PDF Car Sleep
A specialized travel planning agent designed to generate professional HTML and PDF guides specifically for travelers sleeping in their vehicles or EVs.
Claude Code Bridge
A bridge allowing OpenClaw to utilize the local Claude Code CLI for code generation and review without requiring an API key.
yaoyaoya-memory
A sophisticated hierarchical memory system that provides AI assistants with long-term persistence, structured learning logs, and seamless knowledge base synchronization.
Monero Wallet Recovery Guide
A definitive technical guide for restoring access to Monero funds using mnemonic seeds, private keys, and hardware wallets.
Xiaohongshu Creator Automation
A specialized automation tool for Xiaohongshu creators to standardize content publishing and improve post success rates.
Featured*
