Token导航 LogoToken导航TokenDH.com
开发敏感数据clawhub未标认证来源可访问clear审计提醒

telegram-autopilotTelegram autopilot 开发

Agent Skill

telegram-autopilot 用于补充开发相关能力,适合在 OpenClaw 中需要让 Agent 承接开发相关任务时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

8,846

周安装

380

GitHub Stars

公开资料未说明

下载量

3,101
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:telegram-autopilot(Telegram autopilot 开发)
来源仓库:https://github.com/shor73/telegram-autopilot
安装命令:
openclaw skills install telegram-autopilot
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install telegram-autopilot

简介

telegram-autopilot 管理个人账号的自动回复系统,模拟用户身份响应私信。

  • 适用于客服、信息分发或个人助理类自动化场景。
  • 可自定义回复规则与触发条件。telegram-autopilot 属于开发类 Skill,可作为该场景下的辅助能力补充。
  • 安装前需授权账号权限并评估隐私风险。
  • 建议在非工作时间先行测试,避免误发消息。

SKILL.md

name
telegram-autopilot
description
Manage a Telegram userbot autopilot that responds to private messages as the user using AI. Use when the user wants to set up auto-replies on their personal Telegram account, manage allowed contacts, configure AI response style, or send messages/media as themselves. Triggers on "telegram autopilot", "auto reply telegram", "manage my telegram", "respond for me", "telegram userbot", "paid media telegram". Requires secrets — Telegram API credentials (api_id, api_hash), phone number, optional 2FA password, AI provider API key (Anthropic or OpenAI-compatible), optional Telegram bot token for owner notifications.

Telegram Autopilot

Source: https://github.com/Shor73/telegram-autopilot Author: @Shor73 License: MIT

AI-powered autopilot for personal Telegram accounts. Responds to private messages as the user when they're unavailable.

Prerequisites

  • Python 3.10+
  • Telethon: pip3 install telethon
  • Telegram API credentials (api_id + api_hash from https://my.telegram.org)
  • An Anthropic API key (for AI responses)

Setup

1. Get Telegram API Credentials

Direct the user to https://my.telegram.org → API Development Tools → create app. They need: API ID (number) and API Hash (string).

2. Login Flow

Telegram login requires: phone → OTP code → optional 2FA password.

Critical: OTP codes expire in ~60 seconds. Minimize latency:

  • Use a file-based code exchange (script polls a file every 200ms)
  • Or serve a simple web form on a local port for instant code entry
  • Never rely on chat round-trip for code delivery — too slow

Run scripts/setup.py with the user's credentials. It handles:

  1. Requesting the OTP code
  2. Waiting for code via file (enter_code.txt)
  3. 2FA password if needed
  4. Saving the session file
python3 scripts/setup.py --api-id 12345 --api-hash "abc123" --phone "+1234567890"

The session file (.session) persists auth — login is one-time only.

3. Configure Contacts

Edit config.json to define allowed contacts:

{
  "contacts": {
    "username": {
      "name": "Display Name",
      "id": 123456789,
      "tone": "friendly",
      "language": "en"
    }
  },
  "ai": {
    "provider": "anthropic",
    "model": "claude-sonnet-4-6",
    "api_key": "sk-ant-...",
    "max_tokens": 300
  },
  "owner": {
    "name": "Owner Name",
    "bio": "Brief description for the AI persona",
    "telegram_id": 123456789
  },
  "notifications": {
    "bot_token": "optional-bot-token-for-notifications",
    "chat_id": "optional-chat-id"
  }
}

4. Start Autopilot

python3 scripts/autopilot.py --config config.json --session session_name

Features

Auto-Reply

  • Responds only to allowed contacts in private chats
  • Marks messages as read before replying (natural behavior)
  • Simulates typing delay proportional to response length
  • Maintains conversation history for context

AI Persona

  • System prompt configurable per-contact (tone, language)
  • Never invents facts — says "I'll check and get back to you" for unknowns
  • Never reveals it's AI

Notifications

  • Forwards received messages to owner via Telegram bot
  • Forwards sent replies for monitoring

Paid Media (Channels Only)

Telegram paid media (inputMediaPaidMedia) only works in channels, not private chats.

To send paid media:

  1. Create a private channel
  2. Post media with InputMediaPaidMedia(stars_amount=N, extended_media=[...])
  3. Generate invite link and send to recipient
python3 scripts/send_paid_media.py --session session_name --target username --photo /path/to/photo.jpg --stars 1

Management Commands

Stop autopilot: kill the process or send SIGTERM. Add/remove contacts: edit config.json and restart.

Architecture

telegram-autopilot/
├── SKILL.md
├── scripts/
│   ├── setup.py          — Login flow (OTP + 2FA)
│   ├── autopilot.py      — Main event loop
│   ├── send_paid_media.py — Paid media via channel
│   └── code_server.py    — Web form for fast OTP entry
└── references/
    └── telegram-auth.md  — Telegram auth flow documentation

Required Credentials

SecretPurposeWhere used
Telegram API ID + HashMTProto client authsetup.py, autopilot.py
Phone number + 2FAAccount login (one-time)setup.py
AI API keyResponse generationautopilot.py
Bot token (optional)Owner notificationsautopilot.py

All secrets are stored in config.jsonnever commit this file.

Security & Ethics

  • Session security: The .session file grants full access to the account. Protect it like a password.
  • Transparency: The AI is instructed to be honest if directly asked whether it's AI.
  • OTP server: code_server.py binds to 127.0.0.1 (localhost only). Never expose it to the network.
  • Notifications: The skill can forward messages to the owner via bot. Ensure you control the bot token and chat_id.
  • Rate limits: Telegram may restrict accounts with aggressive automation. The autopilot uses natural delays.
  • One session at a time: Only one process can use a session file. Stop autopilot before running other scripts.
  • Platform policies: Using a userbot to auto-reply may violate Telegram ToS. Use at your own risk.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

78.99%
按下载量换算2,449

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills