Run
技能 已验证 活跃One-shot lifecycle command that chains init → baseline → spawn → eval → merge in a single invocation.
Streamline the entire agent development and evaluation workflow by executing a multi-step process in one invocation, saving users time and reducing manual coordination.
功能
- One-shot execution of init → baseline → spawn → eval → merge
- Supports custom agent templates
- Configurable evaluation metrics and directions
- User confirmation for merge step
- Parallel agent spawning
使用场景
- Automating repetitive agent development cycles.
- Ensuring consistent agent evaluation and merging processes.
- Reducing the complexity of managing multi-agent workflows.
- Quickly iterating on agent performance with a single command.
非目标
- Performing individual lifecycle steps in isolation.
- Replacing the detailed configuration of individual AgentHub commands.
- Handling agent failures beyond stopping the workflow and reporting errors.
工作流
- Initialize the agent session with provided parameters.
- Capture a baseline metric if an evaluation command is supplied.
- Spawn agents in parallel, optionally using a specified template.
- Monitor agent completion and display summaries.
- Evaluate results based on metrics or LLM judge mode.
- Confirm merge with the user before executing the merge command.
Scope
- info:Dry-run previewWhile the merge step has a confirmation, a specific `--dry-run` flag for the entire workflow is not explicitly mentioned, though the step-by-step nature provides some preview.
安装
请先添加 Marketplace
/plugin marketplace add alirezarezvani/claude-skills/plugin install agenthub@claude-code-skills质量评分
已验证类似扩展
Agent Worker Specialist
100Agent skill for worker-specialist - invoke with $agent-worker-specialist
Orchestrate
100Wire Commands, Agents, and Skills together for complex features. Use when building features that need research, planning, and implementation phases.
Project Session Manager
100Worktree-first dev environment manager for issues, PRs, and features with optional tmux sessions
Acp Router
100Route plain-language requests for Pi, Claude Code, Cursor, Copilot, OpenClaw ACP, OpenCode, Gemini CLI, Qwen, Kiro, Kimi, iFlow, Factory Droid, Kilocode, or explicit ACP harness work into either OpenClaw ACP runtime sessions or direct acpx-driven sessions ("telephone game" flow). For coding-agent thread requests, read this skill first, then use only `sessions_spawn` for thread creation. Codex chat binding defaults to the native Codex app-server plugin unless ACP is explicit or background spawn needs ACP.
Autopilot Loop
99Run an autonomous /loop iteration -- check progress, work on next task, schedule next wake
Team Skill
99N coordinated agents on shared task list using Claude Code native teams