Telegram ControlSAFE
Remote control AI coding assistants (Claude Code/Codex) via Telegram
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Remote control AI coding assistants (Claude Code / Codex) via Telegram
[](https://pypi.org/project/telegram-mcp-server/) [](https://pypi.org/project/telegram-mcp-server/) [](LICENSE)
English | 简体中文
Why This Project?
Have you ever encountered these scenarios:
- 💤 Late at night in bed, you suddenly think of a bug that needs fixing, but don't want to get up and open your laptop?
- 🚇 On your commute, you want AI to refactor code for you, but your laptop isn't with you?
- 🏢 Multiple Claude Code or Codex sessions running on remote servers, and you want to check their progress anytime?
- ⏰ Long-running tasks (testing, building, refactoring) that take hours, but you don't want to sit in front of the computer?
Telegram MCP Server was created to solve these problems!
Through the MCP (Model Context Protocol), this project allows you to:
- 📱 Anytime, anywhere view and control AI coding assistants via Telegram
- 🔄 Multi-session management: Use
screenon remote servers to manage multiple projects simultaneously - 🌙 True unattended mode: Wait up to 7 days with smart polling, minimal system resources
- 💬 Simple interaction: Send messages via Telegram to give AI assistants next instructions
Perfect for:
- 24/7 remote servers
- Long-running tasks
- Multi-project parallel development
- Remote work from anywhere
Features
- 🌙 True Unattended Mode - Wait up to 7 days with smart progressive polling
- 📱 Remote Control - Control AI assistants from anywhere via Telegram
- 🔄 Two-way Communication - Send notifications, receive replies, continuous dialogue
- 📁 File Operations - View and download project files
- 🎯 Multi-session Management - Manage multiple projects simultane
e4389296fb23OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add telegram-mcp-server --env TELEGRAM_BOT_TOKEN=${TELEGRAM_BOT_TOKEN} -- uvx telegram-mcp-server{
"mcpServers": {
"telegram-mcp-server": {
"command": "uvx",
"args": [
"telegram-mcp-server"
],
"env": {
"TELEGRAM_BOT_TOKEN": "${TELEGRAM_BOT_TOKEN}"
}
}
}
}Exposed tools (9)
5 read · 4 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
telegram_get_context_info | read | |
telegram_notify | read | |
telegram_notify_with_actions | read | |
telegram_send | write | |
telegram_send_code | write | |
telegram_send_file | write | |
telegram_send_image | write | |
telegram_unattended_mode | read | |
telegram_wait_reply | read |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (4 observation(s))
- Network
- declared (6 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (7)
photo_id = hashlib.md5(photo.file_id.encode()).hexdigest()[:8]
action_id = hashlib.md5(
mcp, python-telegram-bot, aiofiles
**Solution 1: Add to PATH**
# Add to ~/.zshrc or ~/.bashrc
curl -fsSL https://raw.githubusercontent.com/batianVolyc/telegram-mcp-server/main/install.sh | bash
curl -LsSf https://astral.sh/uv/install.sh | sh
Gates applied: no_behavioural_pass.
e4389296fb23full audit observations/trust-audit/mcp-server/batianvolyc__telegram-control.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | e4389296fb23 | SAFE | B | 89 | first audit |
Questions
What is the Telegram Control MCP server?
Remote control AI coding assistants (Claude Code/Codex) via Telegram
What tools does Telegram Control expose?
9 in total: 5 read-only, 4 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Telegram Control safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Telegram Control need?
It reads TELEGRAM_BOT_TOKEN from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Telegram Control run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as telegram-mcp-server.
How current is this page?
The grade is for one exact copy of the source (e4389296fb23), read on 2026-10-08. The repository is watched and re-audited when it changes.