Hive Workers Skill
Orchestrates complex task execution using ruflo hive mind with automatic code review and documentation updates.
Quick Start
0. EXPLORE: Search codebase for existing implementations (MANDATORY)
1. Read the task or issue
2. Initialize hive mind: swarm_init({ topology: "hierarchical" })
3. Create todos: TodoWrite with all tasks
4. Execute in parallel where possible
5. Run code review with Task agent
6. Fix any "Must Fix" items
7. Run governance audit
8. Create ADR if architectural decision
9. Update project status
10. Generate sprint report
11. Cleanup: swarm_destroy, mark issues done
Execution
When triggered, immediately:
- Read
agent-prompt.mdfrom this skill's directory (./agent-prompt.md) - Spawn a single Task with
subagent_type: "general-purpose"passing the agent-prompt content as the prompt - Include in the prompt: the user's request, current working directory, and any arguments passed
- Wait for the agent to complete
- Present the agent's summary to the user
Sub-Documentation
| Document | Contents |
|---|---|
| agent-prompt.md | Full execution workflow, phase details, checklists |
| context-persistence.md | Maintaining context across sessions for long-running tasks |
| CHEATSHEET.md | Quick reference for common operations |
| templates/ | Workflow templates for task execution |
Related Skills
- governance - Standards enforcement
- worktree-manager - Parallel development
Prerequisites
- ruflo configured (
claude mcp add ruflo -- npx ruflo@v3alpha mcp start) - Linear MCP (optional, for issue tracking)
For full execution details, see agent-prompt.md.
Changelog
v1.0.0 (2026-03)
- Initial public release
- 11-step hive mind execution workflow
- ruflo MCP integration (
swarm_init,agent_spawn,task_orchestrate,swarm_destroy) - Automatic governance audit after execution
- Context persistence across sessions (
context-persistence.md) - Workflow templates for common task types