telegram (Claude Code official plugins): permissions, hooks, MCP and scripts

All skills & plugins / Claude Code official plugins

Telegram messaging bridge with built-in access control. Manage pairing, allowlists, and policy via /telegram:access.

Type: Plugin with 2 skills
Catalog: Claude Code official plugins
Source: anthropics/claude-plugins-official/external_plugins/telegram (11 files scanned)
Install and details: Agent Skills Index entry
License: Apache-2.0 (LICENSE file)
Signals: Pre-approved shellLocal MCP serverBundled scriptsNetwork calls in codeReads credentialsSpawns processes

Pre-approved tools

Shell (scoped). Tools listed in allowed-tools (or an agent's tools) can be used without a permission prompt while the skill is active, depending on your agent and settings.

Kind	Tools
read/search	`Read`
write/edit files	`Write`
shell, scoped	`Bash(ls )Bash(mkdir )Bash(echo )Bash(chmod )`

Declared in configure/SKILL.md, access/SKILL.md.

Hooks

No hooks.

MCP servers and apps

Server	Where it runs	Credentials
telegram telegram/.mcp.json	Local process: `bun run --cwd ${CLAUDE_PLUGIN_ROOT} --shell=bun --silent start`	`CLAUDE_PLUGIN_ROOT`

Bundled code

1 code file (TypeScript):

external_plugins/telegram/server.ts

Network calls

1 file with a network call pattern. First match per file:

external_plugins/telegram/server.ts:611
```
const res = await fetch(url)
```

Starts other programs

1 file. First match per file:

external_plugins/telegram/server.ts:24

import { execFileSync } from 'child_process'

Credentials referenced

TELEGRAM_BOT_TOKENTOKEN

Static scan of commit 41f619d on 2026-10-27. Signals are patterns found in the files, not a verdict: a network call can be the whole point of a skill. Read the linked lines before you decide. How the scan works.