LeoYeAI/agent-master-skills / skills/agenthub
agenthub
VerifiedUse when a task should be attempted by several agents in parallel and the best result kept (code speed, copy variants, research angles). Spawns N agents in git worktrees, ranks results by metric or judge with result_ranker.py and merges the winner.
AgentHub — Multi-Agent Collaboration
Spawn N parallel AI agents that compete on the same task. Each agent works in an isolated git worktree. The coordinator evaluates results and merges the winner.
Slash Commands
| Command | Description |
|---|---|
/hub-init | Create a new collaboration session — task, agent count, eval criteria |
/agenthub:spawn | Launch N parallel subagents in isolated worktrees |
/hub-status | Show DAG state, agent progress, branch status |
/agenthub:eval | Rank agent results by metric or LLM judge |
/agenthub:merge | Merge winning branch, archive losers |
/agenthub:board | Read/write the agent message board |
/hub-run | One-shot lifecycle: init → baseline → spawn → eval → merge |
Agent Templates
When spawning with --template, agents follow a predefined iteration pattern:
| Template | Pattern | Use Case |
|---|---|---|
optimizer | Edit → eval → keep/discard → repeat x10 | Performance, latency, size |
refactorer | Restructure → test → iterate until green | Code quality, tech debt |
test-writer | Write tests → measure coverage → repeat | Test coverage gaps |
bug-fixer | Reproduce → diagnose → fix → verify | Bug fix approaches |
Templates are defined in references/agent-templates.md.
When This Skill Activates
Trigger phrases:
- "try multiple approaches"
- "have agents compete"
- "parallel optimization"
- "spawn N agents"
- "compare different solutions"
- "fan-out" or "tournament"
- "generate content variations"
- "compare different drafts"
- "A/B test copy"
- "explore multiple strategies"
Coordinator Protocol
The main Claude Code session is the coordinator. It follows this lifecycle:
INIT → DISPATCH → MONITOR → EVALUATE → MERGE
1. Init
Run /hub-init to create a session. This generates:
.agenthub/sessions/{session-id}/config.yaml— task config.agenthub/sessions/{session-id}/state.json— state machine.agenthub/board/— message board channels
2. Dispatch
Run /agenthub:spawn to launch agents. For each agent 1..N:
- Post task assignment to
.agenthub/board/dispatch/ - Spawn via Agent tool with
isolation: "worktree" - All agents launched in a single message (parallel)
3. Monitor
Run /hub-status to check progress:
dag_analyzer.py --status --session {id}shows branch state- Board
progress/channel has agent updates
4. Evaluate
Run /agenthub:eval to rank results:
- Metric mode: run eval command in each worktree, parse numeric result
- Judge mode: read diffs, coordinator ranks by quality
- Hybrid: metric first, LLM-judge for ties
5. Merge
Run /agenthub:merge to finalize:
git merge --no-ffwinner into base branch- Tag losers:
git tag hub/archive/{session}/agent-{i} - Clean up worktrees
- Post merge summary to board
Agent Protocol
Each subagent receives this prompt pattern:
You are agent-{i} in hub session {session-id}.
Your task: {task description}
Instructions:
1. Read your assignment at .agenthub/board/dispatch/{seq}-agent-{i}.md
2. Work in your worktree — make changes, run tests, iterate
3. Commit all changes with descriptive messages
4. Write your result summary to .agenthub/board/results/agent-{i}-result.md
5. Exit when done
Agents do NOT see each other's work. They do NOT communicate with each other. They only write to the board for the coordinator to read.
DAG Model
Branch Naming
hub/{session-id}/agent-{N}/attempt-{M}
- Session ID: timestamp-based (
YYYYMMDD-HHMMSS) - Agent N: sequential (1 to agent-count)
- Attempt M: increments on retry (usually 1)
Frontier Detection
Frontier = branch tips with no child branches. Equivalent to AgentHub's "leaves" query.
python scripts/dag_analyzer.py --frontier --session {id}
Immutability
The DAG is append-only:
- Never rebase or force-push agent branches
- Never delete commits (only branch refs after archival)
- Every approach preserved via git tags
Message Board
Location: .agenthub/board/
Channels
| Channel | Writer | Reader | Purpose |
|---|---|---|---|
dispatch/ | Coordinator | Agents | Task assignments |
progress/ | Agents | Coordinator | Status updates |
results/ | Agents + Coordinator | All | Final results + merge summary |
Post Format
---
author: agent-1
timestamp: 2026-03-17T14:30:22Z
channel: results
parent: null
---
## Result Summary
- **Approach**: Replaced O(n²) sort with hash map
- **Files changed**: 3
- **Metric**: 142ms (baseline: 180ms, delta: -38ms)
- **Confidence**: High — all tests pass
Board Rules
- Append-only: never edit or delete posts
- Unique filenames:
{seq:03d}-{author}-{timestamp}.md - YAML frontmatter required on all posts
Evaluation Modes
Metric-Based
Best for: benchmarks, test pass rates, file sizes, response times.
python scripts/result_ranker.py --session {id} \
--eval-cmd "pytest bench.py --json" \
--metric p50_ms --direction lower
The ranker runs the eval command in each agent's worktree directory and parses the metric from stdout.
LLM Judge
Best for: code quality, readability, architecture decisions.
The coordinator reads each agent's diff (git diff base...agent-branch) and ranks by:
- Correctness (does it solve the task?)
- Simplicity (fewer lines changed preferred)
- Quality (clean execution, good structure)
Hybrid
Run metric first. If top agents are within 10% of each other, use LLM judge to break ties.
Session Lifecycle
init → running → evaluating → merged
→ archived (if no winner)
State transitions managed by session_manager.py:
| From | To | Trigger |
|---|---|---|
init | running | /agenthub:spawn completes |
running | evaluating | All agents return |
evaluating | merged | /agenthub:merge completes |
evaluating | archived | No winner / all failed |
Proactive Triggers
The coordinator should act when:
| Signal | Action |
|---|---|
| All agents crashed | Post failure summary, suggest retry with different constraints |
| No improvement over baseline | Archive session, suggest different approaches |
| Orphan worktrees detected | Run session_manager.py --cleanup {id} |
Session stuck in running | Check board for progress, consider timeout |
Installation
# Copy to your Claude Code skills directory
cp -r engineering/agenthub ~/.claude/skills/agenthub
# Or install via ClawHub
clawhub install agenthub
Scripts
| Script | Purpose |
|---|---|
hub_init.py | Initialize .agenthub/ structure and session |
dag_analyzer.py | Frontier detection, DAG graph, branch status |
board_manager.py | Message board CRUD (channels, posts, threads) |
result_ranker.py | Rank agents by metric or diff quality |
session_manager.py | Session state machine and cleanup |
Related Skills
- autoresearch-agent — Single-agent optimization loop (use AgentHub when you want N agents competing)
- self-improving-agent — Self-modifying agent (use AgentHub when you want external competition)
- git-worktree-manager — Git worktree utilities (AgentHub uses worktrees internally)
More in this skill
Load these only when the task needs them.
-
references/coordination-strategies.md — Multi-Agent Coordination Strategies
-
references/dag-patterns.md — Git DAG Patterns for Multi-Agent Collaboration
-
python3 scripts/dry_run.py --help— Dry-run validation for the AgentHub plugin
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 470k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |