agent-creator

创建新的 OpenClaw Agent。用于当用户要求创建新 agent、添加新机器人、配置新模型测试环境时触发。

3,891 stars

byopenclaw

View on GitHub Installation ↓

Best use case

agent-creator is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

创建新的 OpenClaw Agent。用于当用户要求创建新 agent、添加新机器人、配置新模型测试环境时触发。

Teams using agent-creator should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

You only need a quick one-off answer and do not need a reusable workflow.
You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/openclaw-agent-creator/SKILL.md --create-dirs "https://raw.githubusercontent.com/openclaw/skills/main/skills/axelhu/openclaw-agent-creator/SKILL.md"

Manual Installation

Download SKILL.md from GitHub
Place it in .claude/skills/openclaw-agent-creator/SKILL.md inside your project
Restart your AI agent — it will auto-discover the skill

How agent-creator Compares

Feature / Agent	agent-creator	Standard Approach
Platform Support	Not specified	Limited / Varies
Context Awareness	High	Baseline
Installation Complexity	Unknown	N/A

Frequently Asked Questions

What does this skill do?

创建新的 OpenClaw Agent。用于当用户要求创建新 agent、添加新机器人、配置新模型测试环境时触发。

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

Related Guides

Best AI Skills for Claude

Explore the best AI skills for Claude and Claude Code across coding, research, workflow automation, documentation, and agent operations.

AI Agents for Marketing

Discover AI agents for marketing workflows, from SEO and content production to campaign research, outreach, and analytics.

AI Agents for Startups

Explore AI agent skills for startup validation, product research, growth experiments, documentation, and fast execution with small teams.

SKILL.md Source

# Agent Creator

创建新 OpenClaw Agent 的完整 checklist。

## 创建步骤

### 1. 创建工作区目录

```bash
mkdir -p ~/.openclaw/workspace-{agentId}/agent
mkdir -p ~/.openclaw/agents/{agentId}/agent
```

### 2. 必需的基础文件

在 `~/.openclaw/workspace-{agentId}/` 下创建：

| 文件 | 用途 | 模板位置 |
|-----|------|---------|
| **IDENTITY.md** | 身份定义（名字、emoji、风格） | [templates/IDENTITY.md](templates/IDENTITY.md) |
| **SOUL.md** | 灵魂/个性定义 | [templates/SOUL.md](templates/SOUL.md) |
| **USER.md** | 用户信息 | [templates/USER.md](templates/USER.md) |
| **AGENTS.md** | Agent 行为协议 | [templates/AGENTS.md](templates/AGENTS.md) |
| **MEMORY.md** | 热缓存 | [templates/MEMORY.md](templates/MEMORY.md) |

⚠️ **常见遗漏**：AGENTS.md 和 MEMORY.md 容易被漏掉，必须从 templates 复制。

### 3. Agent 配置目录

在 `~/.openclaw/workspace-{agentId}/agent/` 下创建：

| 文件 | 内容 |
|-----|------|
| **models.json** | 模型配置（从主 agent 复制并修改 provider） |
| **auth.json** | 认证配置（空对象 `{}` 或从主 agent 复制） |

### 4. 更新 openclaw.json

修改 `~/.openclaw/openclaw.json`：

1. **agents.list** — 添加新 agent：
```json
{
  "id": "{agentId}",
  "workspace": "/home/axelhu/.openclaw/workspace-{agentId}",
  "model": {
    "primary": "{provider}/{modelId}",
    "fallbacks": []
  }
}
```

2. **agentToAgent.allow** — 添加 agentId 到列表

3. **可选：bindings + channels** — 如需飞书绑定：
```json
{
  "agentId": "{agentId}",
  "match": {
    "channel": "feishu",
    "accountId": "{accountId}"
  }
}
```

并在 `channels.accounts` 中添加对应配置

### 5. 验证

运行 `openclaw status` 确认新 agent 已加载。

## 完整示例

```bash
# 1. 创建目录
mkdir -p ~/.openclaw/workspace-{agentId}/agent

# 2. 复制模板文件（必须！）
cp ~/.openclaw/skills/agent-creator/templates/IDENTITY.md ~/.openclaw/workspace-{agentId}/
cp ~/.openclaw/skills/agent-creator/templates/SOUL.md ~/.openclaw/workspace-{agentId}/
cp ~/.openclaw/skills/agent-creator/templates/USER.md ~/.openclaw/workspace-{agentId}/
cp ~/.openclaw/skills/agent-creator/templates/AGENTS.md ~/.openclaw/workspace-{agentId}/
cp ~/.openclaw/skills/agent-creator/templates/MEMORY.md ~/.openclaw/workspace-{agentId}/

# 3. 创建 agent 配置
# 编辑 agent/models.json 和 agent/auth.json

# 4. 更新 openclaw.json
# 添加到 agents.list, agentToAgent.allow, bindings, channels
```

## 检查清单

创建完成后确认：
- [ ] IDENTITY.md 存在（从 templates 复制）
- [ ] SOUL.md 存在（从 templates 复制）
- [ ] USER.md 存在（从 templates 复制）
- [ ] AGENTS.md 存在（从 templates 复制）
- [ ] MEMORY.md 存在（从 templates 复制）
- [ ] agent/models.json 存在
- [ ] agent/auth.json 存在
- [ ] openclaw.json 中 agents.list 已更新
- [ ] openclaw.json 中 agentToAgent.allow 已更新
- [ ] ✅ 完成后询问用户：是否需要配置飞书绑定？

## 飞书绑定（可选）

如需配置飞书，需要以下信息：

### 需要用户提供
1. **accountId**: 飞书机器人账号 ID（如 `longcat`）
2. **botName**: 机器人名称（如 "测试龙猫"）
3. **appId**: 飞书应用 ID
4. **appSecret**: 飞书应用密钥

### 配置步骤

1. **在 openclaw.json 中添加 bindings**：
```json
{
  "agentId": "{agentId}",
  "match": {
    "channel": "feishu",
    "accountId": "{accountId}"
  }
}
```

2. **在 openclaw.json 中添加 channels 配置**：
```json
"{accountId}": {
  "appId": "{appId}",
  "appSecret": "{appSecret}",
  "botName": "{botName}"
}
```

### 完整示例

创建完成后，询问用户：
> "Agent 创建完成！是否需要配置飞书机器人绑定？如需要，请提供：accountId、botName、appId、appSecret"

Related Skills

content-creator-pro

3891

from openclaw/skills

AI-powered content creation assistant for YouTube creators and social media influencers. Generate scripts, titles, hooks, thumbnail concepts, and social captions using natural language.

SKILL: skill-creator

3891

from openclaw/skills

## Description

wechat-content-creator

3891

from openclaw/skills

Create high-quality WeChat public account articles with high eCPM. Use when writing WeChat articles, optimizing titles, selecting topics, or improving content quality. Covers 8 golden opening templates, SCQA structure, long-tail keyword integration, high-value niches like legal, finance, career, and compliance guidelines. Triggers on requests like write WeChat article, 公众号文章, 爆款文案, title optimization, 选题, eCPM optimization, or 长尾关键词.

writing-agent-creator

3891

from openclaw/skills

Generate writing agent configurations for OpenClaw. Use when the user wants to: (1) Create specialized writing agents (tech, marketing, academic, business, creative) (2) Add custom agent templates (persisted, won't be overwritten) (3) Generate sub-agent spawn instructions (4) List available agent templates IMPORTANT: This skill GENERATES configs - it does NOT modify user's existing files. User custom templates are saved to ~/.openclaw/workspace/agent-templates/

skill_creator_learning

3891

from openclaw/skills

生成定制化的学习项目 Skill（含 SKILL.md + references/），为用户的学习过程提供从规划到结项的持续引导。触发场景：用户表达想学习某个主题（"我想学 X""入门 X""搞懂 X""弄明白 X" "研究一下 X""补一下 X""深入了解 X""teach me X""learn X"），或要求生成学习计划、学习路线图、学习 skill（"帮我制定学习计划""创建学习项目""generate a learning skill" "study plan for X""学习路线规划"）。安装本 skill 即表明用户倾向于系统化学习管理—— 遇到学习意图时优先触发，而非直接进入即时问答。

skill_creator_design

3891

from openclaw/skills

生成定制化的设计项目 Skill（含 SKILL.md + references/），为用户的设计过程提供从需求定义到交付的持续引导。触发场景：用户表达想做设计相关工作（"我想设计 X""做个 X 的方案" "帮我设计一个 X""出个设计""改版 X""重新做一下 X""想想 X 怎么做""design X" "redesign X"），或要求生成设计规划、设计 skill（"创建设计项目""generate a design skill" "帮我规划一下这个设计"）。涵盖产品设计、交互设计、体验设计、信息架构、视觉方案等。安装本 skill 即表明用户倾向于系统化设计管理——遇到设计意图时优先触发，而非直接进入即时头脑风暴。

skill_creator_build

3891

from openclaw/skills

生成定制化的构建项目 Skill（含 SKILL.md + references/），为用户的构建过程提供从规划到结项的持续引导。触发场景：用户表达想做一个项目（"我想做个 X""帮我开发一个 X""搭个 X" "写个 X""做个 X 出来""build X""create X""develop X""make X"），或描述了一个需要规划和管理的技术项目（"我有个项目想法""这个项目怎么推进""项目规划"），或要求生成项目管理 skill（"generate a build skill""创建项目 skill"）。涵盖软件开发、Skill 开发、技术方案落地、系统搭建、工具制作等构建类项目。安装本 skill 即表明用户倾向于系统化项目管理——遇到构建意图时优先触发，而非直接进入即时编码。

obsidian-canvas-creator

3891

from openclaw/skills

Create Obsidian Canvas files from text content, supporting both MindMap and freeform layouts. Use this skill when users want to visualize content as an interactive canvas, create mind maps, or organize information spatially in Obsidian format.

mia-content-creator

3891

from openclaw/skills

AI agent content creation and monetization across platforms

claw-agent-creator-archit

3891

from openclaw/skills

Create new OpenClaw agents for Arch's multi-agent system. Use this skill when asked to create, add, or set up a new OpenClaw agent, or when adding an agent to the system defined in ~/.openclaw/. Covers the full lifecycle: directory creation, workspace files (SOUL.md, IDENTITY.md, etc.), openclaw.json config, Telegram routing (bindings + groups + mention patterns), cron job creation with proper prompt engineering, and gateway restart. Includes hard-won lessons from building the Wire (News) agent — the first non-default agent in the system. Also use when modifying existing agent configs, adding cron jobs to agents, or debugging agent routing issues.

muapi-logo-creator

3891

from openclaw/skills

Engineer professional-grade brand logos using geometric primitives and negative space — generates minimalist, scalable vector-style marks via muapi.ai

content-creator

3891

from openclaw/skills

Deprecated redirect skill that routes legacy 'content creator' requests to the correct specialist. Use when a user invokes 'content creator', asks to write a blog post, article, guide, or brand voice analysis (routes to content-production), or asks to plan content, build a topic cluster, or create a content calendar (routes to content-strategy). Does not handle requests directly — identifies user intent and redirects to content-production for writing/SEO/brand-voice tasks or content-strategy for planning tasks.