スキル agent-manager-skill
📦

agent-manager-skill

安全

Manage CLI Agents with Tmux

This skill enables Claude to manage multiple local CLI agents running in parallel tmux sessions, providing start/stop/monitor capabilities and cron-friendly scheduling for automated workflows.

対応: Claude Codex Code(CC)
🥉 74 ブロンズ
1

スキルZIPをダウンロード

2

Claudeでアップロード

設定 → 機能 → スキル → スキルをアップロードへ移動

3

オンにして利用開始

テストする

「agent-manager-skill」を使用しています。 agent-manager-skill list

期待される結果:

  • EMP_0001: running (uptime: 2h 34m)
  • EMP_0002: running (uptime: 1h 12m)
  • EMP_0003: stopped

「agent-manager-skill」を使用しています。 agent-manager-skill doctor

期待される結果:

  • ✓ tmux installed (v3.3a)
  • ✓ python3 installed (3.10.12)
  • ✓ agents/ directory exists
  • ✓ 5 agents configured
  • All checks passed

セキュリティ監査

安全
v1 • 2/24/2026

Security audit complete. All 9 static findings are false positives. The skill contains only documentation with markdown code examples - no actual shell execution, no suspicious network activity, and no cryptographic code. The backticks flagged are markdown formatting, and the URL is a legitimate GitHub repo reference.

1
スキャンされたファイル
43
解析された行数
0
検出結果
1
総監査数
セキュリティ問題は見つかりませんでした
監査者: claude

品質スコア

38
アーキテクチャ
100
保守性
85
コンテンツ
50
コミュニティ
100
セキュリティ
91
仕様準拠

作れるもの

DevOps Workflow Automation

Run multiple background maintenance scripts in parallel, monitor their output, and schedule recurring tasks using cron.

Parallel AI Task Execution

Launch multiple Claude or other CLI agents simultaneously for different subtasks, then aggregate results.

Background Job Management

Start long-running processes in tmux sessions that persist beyond terminal sessions, with easy monitoring.

これらのプロンプトを試す

List All Agents 
Use the agent-manager-skill to list all currently running agents and their status.
Start Specific Agent 
Start agent EMP_0001 using the agent-manager-skill and show me the output.
Monitor Agent Logs 
Monitor the logs of agent EMP_0002 in real-time using the agent-manager-skill.
Assign Task to Agent 
Assign the following task to agent EMP_0003 using the agent-manager-skill:

Follow teams/fractalmind-ai-maintenance.md Workflow

ベストプラクティス

  • Configure agents in the agents/ directory before attempting to start them
  • Use the doctor command first to verify all prerequisites are installed
  • Monitor agents during initial runs to ensure they behave as expected
  • Use descriptive agent IDs (like EMP_0001) for easy identification

回避

  • Starting agents without first running the doctor check
  • Assigning tasks to agents that are not properly configured
  • Not monitoring agent output during critical operations
  • Running too many agents simultaneously without resource monitoring

よくある質問

What prerequisites does this skill require?
This skill requires tmux and python3 to be installed on your system. Run 'python3 agent-manager/scripts/main.py doctor' to verify installation.
How do I configure new agents?
Place agent configuration files in the agents/ directory. Refer to the repository examples for the required format and structure.
Can I run this skill remotely over SSH?
Yes, as long as tmux is installed on the remote system. The skill manages local tmux sessions and does not require a GUI.
How do I stop a running agent?
Use the stop command with the agent ID: 'python3 agent-manager/scripts/main.py stop EMP_0001'
Is this skill secure?
The skill runs local CLI commands only. Review agent configurations before running untrusted code. No network access is required by default.
Can I schedule agents to run on a cron schedule?
Yes, the skill supports cron-friendly scheduling. Configure scheduled tasks in the agents/ directory with cron expressions.

開発者の詳細

作成者

sickn33

ライセンス

MIT

リポジトリ

https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/agent-manager-skill

参照

main

ファイル構成

📄 SKILL.md