Claude Code Agents Library
Browse 5002 Claude Code subagents that specialize in specific tasks — delegate work to purpose-built AI assistants.
Claude Code subagents are markdown-defined agents stored in .claude/agents/. Each subagent has its own system prompt, tool access, and model — invoke it from the main Claude session with @agent-name to delegate specialized work.
qiaomu-goal-meta-skill
NewTurn vague or complex Codex tasks into strong `/goal` commands with outcome, verification, constraints, boundaries, iteration policy, completion evidence, and pause/block conditions. Use when the user asks for Codex goal instructions, Goal 指令, 目标指令, `/goal` prompts, 中文 Goal 模板, plan-to-goal interviews, success criteria, verification commands, or bounded agent work definitions.
maestro-flow-one
NewAll Maestro workflow commands as a single Claude Code skill — intent routing, decision gates, minimal closed-loop chains
agent-sdk-dev
Development kit for working with the Claude Agent SDK
agentforce-adlc
Agentforce Agent Development Life Cycle — author, discover, scaffold, deploy, test, and optimize .agent files
fanbox
Newvibe coding 的驾驶舱:左边文件,右边/下边终端,中间看清每一次改动。 / The cockpit for vibe coding: browse files on the left, command agents on the right, watch every change in between.
skills
NewSkills for coding agents
luban-skill
New鲁班 | Luban — 把'能用的Skill'打磨成'能被装、能传播、能验证、能进化'的公共资产。Agent skill-polishing workshop: 验料·访行·过尺·慢刨·回炉
@donedev/claude-skill
NewDone Agent Pack — Claude Code skill. Drop SKILL.md into ~/.claude/skills/done/ to teach Claude Code how to operate Done workspaces natively.
@beehive-xr/xr-skill
NewCodex and Claude Code skill installer for Beehive XR workflows
compass-skills
New司南:个性化 AI 任务总控 Skills 系统 /COMPASS: Personal Alignment Skills OS for AI Agents
hermes-agent
NewThe agent that grows with you
@leokon3/agentflow
NewInstall the AgentFlow Claude Code skill for role-based agent workflows.
Getting Started
mkdir -p .claude/agents
Save a markdown file to .claude/agents/agent-name.md with YAML frontmatter (name, description, tools, model).
In a Claude Code session, mention the agent by name to delegate work to it.
Frequently Asked Questions
What are Claude Code subagents?
Subagents are specialized AI assistants defined in .claude/agents/*.md. Each has its own system prompt, allowed tools, and model selection. You delegate work to them from the main Claude session by mentioning their name.
How do I create a subagent?
Create a markdown file in .claude/agents/ with YAML frontmatter specifying name, description, and optionally tools and model. The filename (minus .md) becomes the agent's invocation name.
How are subagents different from skills?
Skills are prompt templates invoked as slash commands for one-shot guidance. Subagents are persistent specialized contexts you delegate ongoing work to via @agent-name. Skills tell Claude what to do once; subagents become a collaborator for the duration of a task.
Where does this agents data come from?
This directory syncs subagents from anthropics/claude-plugins-official, community registries, and curated GitHub collections. Data is updated daily.