isBanned API — Telegram ban checker for LLM agents & bots

Check whether a Telegram channel, group, bot or user is banned, restricted or shadowbanned — with a single HTTP call. No SDK, no key for the free tier. Built for Function Calling / tool use in LangChain, n8n, Custom GPTs, MCP clients and plain bots.

• Website: https://isbanned.app
• OpenAPI: https://isbanned.app/openapi.json
• llms.txt: https://isbanned.app/llms.txt
• MCP server: isbanned-mcp (Claude Desktop / Cursor / Cline)

Quick start — no key needed

curl "https://isbanned.app/check?url=https://t.me/durov"
# or by username:
curl "https://isbanned.app/check?channel=durov"

{
  "link": "https://t.me/durov",
  "status_id": 0,
  "status_key": "active",
  "status_label": "Active",
  "title": "Pavel Durov",
  "banned": false,
  "ok": true
}

banned is the quick yes/no. The free public endpoint is rate-limited to ~5 requests/day per IP. For higher limits, batch (up to 20 links) and deep checks (subscribers, datacenter, age, shadowban), get a personal key from @IsBannedBot (/api) and use https://isbanned.app/<key>-<slug>.

Status codes

id	meaning
0	active — no restriction
1	porn-ban — real Telegram restriction for 18+/porn content (hidden in official iOS/Android apps & some countries, opens on web/desktop)
2	invalid link
3	Sensitive (18+) — soft interface age-gate, not a ban
4 / 5	blocked on iOS only / Android only
6	SCAM / FAKE label
7	removed (Terms of Service)
8	copyright / DMCA
9	restricted
10	invite link requiring a join request

Examples

• examples/check.py — Python (requests)
• examples/langchain_tool.py — LangChain tool for Function Calling
• examples/node.mjs — Node.js (fetch)
• examples/n8n.md — n8n HTTP Request node
• mcp/ — Model Context Protocol server

Custom GPT

Build a "Telegram Ban Checker" GPT in 2 minutes: import the action from https://isbanned.app/openapi-gpt.json (auth: none). See CUSTOM_GPT.md.

MIT