Agenthub
Plugin Verified ActiveMulti-agent collaboration plugin for Claude Code. Spawn N parallel subagents that compete on code optimization, content drafts, research approaches, or any problem that benefits from diverse solutions. Evaluate by metric or LLM judge, merge the winner. 7 slash commands, agent templates, git DAG orchestration, message board coordination.
To enable users to leverage multiple AI agents simultaneously to explore diverse solutions to a problem, evaluate their outcomes, and integrate the best result, significantly accelerating complex tasks like optimization or content creation.
Features
- Spawn N parallel agents in isolated Git worktrees
- Orchestrate multi-agent workflows (init, spawn, eval, merge)
- Evaluate results via configurable metrics or LLM judge
- Merge best agent's work into base branch
- Support for agent templates (optimizer, refactorer, etc.)
- Git DAG orchestration and message board coordination
Use Cases
- Optimize code performance by trying multiple approaches in parallel
- Generate and compare competing content drafts (blog posts, marketing copy)
- Explore diverse research strategies or solutions to a problem
- Automate bug fixing by having agents compete on diagnosis and resolution
Non-Goals
- Directly executing arbitrary shell commands without a defined task
- Replacing core Git functionality
- Managing agent-specific dependencies outside of the worktree context
- Performing tasks unrelated to multi-agent collaboration or optimization
Installation
First, add the marketplace
/plugin marketplace add alirezarezvani/claude-skills/plugin install agenthub@claude-code-skillsContains 8 extensions
Skill (8)
Multi-agent collaboration plugin that spawns N parallel subagents competing on the same task via git worktree isolation. Agents work independently, results are evaluated by metric or LLM judge, and the best branch is merged. Use when: user wants multiple approaches tried in parallel — code optimization, content variation, research exploration, or any task that benefits from parallel competition. Requires: a git repo.
Read, write, and browse the AgentHub message board for agent coordination.
Evaluate and rank agent results by metric or LLM judge for an AgentHub session.
Create a new AgentHub collaboration session with task, agent count, and evaluation criteria.
Merge the winning agent's branch into base, archive losers, and clean up worktrees.
One-shot lifecycle command that chains init → baseline → spawn → eval → merge in a single invocation.
Launch N parallel subagents in isolated git worktrees to compete on the session task.
Show DAG state, agent progress, and branch status for an AgentHub session.
Quality Score
VerifiedTrust Signals
Similar Extensions
Oh My Claudecode
99Multi-agent orchestration system for Claude Code
Ag2 Agent Builder
100Build AG2 (AutoGen) multi-agent systems with slash commands: scaffold agents, wire workflows, create tools, and review code
Worktrunk
99Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows. This plugin provides configuration guidance (LLM commit messages, project hooks, worktree paths) and automatic activity tracking (🤖/💬 indicators in `wt list` showing active Claude sessions).
Agent Triforce
99Multi-agent development system with three specialized agents (PM, Dev, QA) coordinated through checklist methodology based on The Checklist Manifesto and Boeing's checklist engineering
Multi Platform Apps
99Cross-platform application development coordinating web, iOS, Android, and desktop implementations
Context Management
98Context persistence, restoration, and long-running conversation management