← 返回 Skills 市场
4t-shirt

CreateTelegramClawAgent

作者 Woo, Hedy · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
230
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install create-telegram-agent
功能描述
Create a new Telegram agent in OpenClaw with proper configuration. Use when user wants to create a new Telegram bot agent, including setting up agent workspa...
使用说明 (SKILL.md)

Create Telegram Agent Skill

This skill helps create a new Telegram agent with complete configuration.

Prerequisites

Before starting, ensure you have:

  1. A Telegram bot token (from @BotFather)
  2. The agent's purpose/description
  3. The desired agent name

Workflow

Step 1: Collect Required Information

Ask the user for:

  1. Agent Name (required)

    • Must be unique
    • Use lowercase letters, digits, and hyphens
    • Examples: healthman, news-bot, task-manager
  2. Agent Purpose (required)

    • What will this agent do?
    • Examples: "Personal health coach", "Daily news summarizer", "Task reminder assistant"
  3. Telegram Bot Token (required)

    • Format: 1234567890:ABCdefGHIjklMNOpqrSTUvwxyz
    • Get from @BotFather on Telegram

    If user doesn't have a token, provide these steps:

    How to Create a Telegram Bot Token:

    1. Open Telegram and search for @BotFather
    2. Start a chat with BotFather and send /newbot
    3. Follow the prompts:
      • Enter a name for your bot (display name, e.g., "CookMaster")
      • Enter a username for your bot (must end in 'bot', e.g., "cookmaster_bot")
    4. BotFather will send you a message containing your HTTP API token
    5. Copy the token (format: 1234567890:ABCdefGHIjklMNOpqrSTUvwxyz)

    Important: Keep your token secure. Anyone with your token can control your bot.

Step 2: Generate Agent Configuration

Based on user input, prepare the following configurations:

Agent Definition

{
  "id": "\x3Cagent-id>",
  "name": "\x3CAgent Name>",
  "workspace": "/Users/\x3Cuser>/.openclaw/workspace-\x3Cagent-id>",
  "agentDir": "/Users/\x3Cuser>/.openclaw/agents/\x3Cagent-id>/agent"
}

Telegram Binding

{
  "agentId": "\x3Cagent-id>",
  "match": {
    "channel": "telegram",
    "accountId": "\x3Cbot-id>"
  }
}

Telegram Bot Account

"\x3Cbot-id>": {
  "enabled": true,
  "dmPolicy": "pairing",
  "botToken": "\x3Cfull-bot-token>",
  "groupPolicy": "allowlist",
  "streamMode": "partial"
}

Step 3: Present Configuration for Review

CRITICAL: DO NOT MODIFY openclaw.json DIRECTLY

Present the following to the user in a clear format:


📋 Configuration Summary

The following changes will be made to openclaw.json:

1. Agent Definition (add to agents.list)

{
  "id": "\x3Cagent-id>",
  "name": "\x3CAgent Name>",
  "workspace": "/Users/\x3Cuser>/.openclaw/workspace-\x3Cagent-id>",
  "agentDir": "/Users/\x3Cuser>/.openclaw/agents/\x3Cagent-id>/agent"
}

2. Telegram Binding (add to bindings)

{
  "agentId": "\x3Cagent-id>",
  "match": {
    "channel": "telegram",
    "accountId": "\x3Cbot-id>"
  }
}

3. Telegram Bot Account (add to channels.telegram.accounts)

"\x3Cbot-id>": {
  "enabled": true,
  "dmPolicy": "pairing",
  "botToken": "\x3Cfull-bot-token>",
  "groupPolicy": "allowlist",
  "streamMode": "partial"
}

4. Agent Responsibilities (suggested for AGENTS.md)

Based on the provided purpose, generate a responsibility description covering:

  • Core functions of the agent
  • Working methodology
  • Relationship with the user

Please confirm to proceed with the modifications. Any adjustments needed?

Step 4: Wait for User Confirmation

Only proceed after user explicitly confirms (e.g., "confirm", "proceed", "execute", "go ahead").

Step 5: Execute Configuration

Once confirmed:

  1. Create directories:

    mkdir -p /Users/\x3Cuser>/.openclaw/workspace-\x3Cagent-id>
    mkdir -p /Users/\x3Cuser>/.openclaw/agents/\x3Cagent-id>/agent
    
  2. Update openclaw.json:

    • Add agent to agents.list
    • Add binding to bindings
    • Add bot account to channels.telegram.accounts
  3. Create agent files:

    • AGENTS.md - Work responsibilities and workflow
    • SOUL.md - Agent personality and values
  4. Report completion:

    • Show summary of what was created
    • Remind user to restart OpenClaw gateway (if applicable)
    • Provide next steps (e.g., test the bot)

Important Rules

  1. Never modify openclaw.json without explicit user confirmation
  2. Always show the complete configuration changes before applying
  3. Extract bot ID from token (the number before the colon)
  4. Create all necessary directories before writing files
  5. Generate appropriate AGENTS.md and SOUL.md based on the agent's purpose
安全使用建议
This skill appears to do what it says, but review these items before using it: - Confirm who you trust: the skill source is unknown, so ensure you trust the instructions before giving it a bot token. - Backup openclaw.json first (make a copy) so you can restore if the file is corrupted. - The SKILL.md uses a hard-coded /Users/<user>/... path; ensure it will write to the correct home directory on your OS (use $HOME or the platform-appropriate path), otherwise update the paths. - The skill will write your Telegram bot token into openclaw.json in plain text. If you proceed, restrict file permissions (e.g., chmod 600) and consider whether your environment requires secret encryption. - Ask the skill (or implement) to validate inputs and check for existing agent id/name collisions before modifying config. - Clarify the apparent contradiction: it says 'DO NOT MODIFY openclaw.json DIRECTLY' but then lists steps to update openclaw.json — confirm whether the skill will make changes itself after explicit confirmation or only provide instructions for you to copy. - Prefer atomic edits and a backup/verify step: write to a temp file and validate JSON before replacing the live openclaw.json. - After applying changes, follow the recommended restart/test steps and verify the agent appears in OpenClaw and the bot token functions as expected. If you want, provide the exact home path you use and any existing openclaw.json contents (or a copy) and I can point out exactly what lines would be added and produce a safe, validated patch you can apply manually.
功能分析
Type: OpenClaw Skill Name: create-telegram-agent Version: 1.0.0 The skill bundle provides a structured workflow for an AI agent to assist users in configuring a Telegram bot within the OpenClaw environment. It includes explicit safety instructions (SKILL.md) requiring user confirmation before modifying local configuration files (openclaw.json) and handles sensitive information like Telegram Bot Tokens locally without any evidence of data exfiltration or unauthorized execution.
能力评估
Purpose & Capability
Name/description (create a Telegram agent) align with the instructions: collecting a token, generating agent configuration, creating workspace/agent directories, and updating openclaw.json. The requested actions are expected for this purpose and there are no unrelated credential or binary requests.
Instruction Scope
Instructions are narrowly scoped to agent setup (collect inputs, prepare JSON snippets, create directories, update openclaw.json, write AGENTS.md and SOUL.md). Two practical concerns: the skill assumes a macOS-style path (/Users/<user>/...) rather than using the actual HOME or platform-agnostic paths, and it will write the bot token into openclaw.json (plain-text secret storage) without describing protection, backups, validation, or atomic update. Also the SKILL.md contains a confusing line 'CRITICAL: DO NOT MODIFY openclaw.json DIRECTLY' alongside steps that update openclaw.json after confirmation — clarify whether the skill will or will not perform the modification itself.
Install Mechanism
Instruction-only skill with no install spec and no code files; nothing is downloaded or installed, which minimizes risk.
Credentials
No environment variables or external credentials are requested by the skill itself (the user supplies the Telegram token). Storing the provided bot token in openclaw.json is expected for this feature but is a sensitive operation; the skill does not mention securing the token, restricting file permissions, or backing up the config.
Persistence & Privilege
always is false and the skill is user-invocable only. The skill writes files only within the user's OpenClaw config/workspace paths (as described). It does not request system-wide privileges or modify other skills.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install create-telegram-agent
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /create-telegram-agent 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of create-telegram-agent skill: - Guides users through creating a new Telegram agent in OpenClaw, including workspace setup and token configuration. - Collects agent name, purpose, and Telegram bot token (with instructions for obtaining a token). - Prepares all configuration entries for user review before any changes are made. - Requires explicit user confirmation before modifying openclaw.json. - Generates agent workspace, required directories, AGENTS.md, and SOUL.md files. - Provides a clear summary of actions and next steps upon completion.
元数据
Slug create-telegram-agent
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

CreateTelegramClawAgent 是什么?

Create a new Telegram agent in OpenClaw with proper configuration. Use when user wants to create a new Telegram bot agent, including setting up agent workspa... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 230 次。

如何安装 CreateTelegramClawAgent?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install create-telegram-agent」即可一键安装,无需额外配置。

CreateTelegramClawAgent 是免费的吗?

是的,CreateTelegramClawAgent 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

CreateTelegramClawAgent 支持哪些平台?

CreateTelegramClawAgent 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 CreateTelegramClawAgent?

由 Woo, Hedy(@4t-shirt)开发并维护,当前版本 v1.0.0。

💬 留言讨论