tgden
Public Telegram catalog MCP server: search 1.4M+ channels, groups, bots, and posts, and browse marketplace listings, with no auth or API key.
README
tgden — Telegram catalog as an MCP server + free API
A live index of what exists on Telegram: 1.1M+ channels, 181k+ group chats, 142k+ bots, 4.5M+ indexed posts. No Telegram account. No API key. No install.
- MCP endpoint —
https://tgden.com/api/mcp - REST API —
https://tgden.com/api/catalog - Human UI — tgden.com · Docs — tgden.com/en/api-docs
Why another Telegram MCP server?
Every other Telegram MCP server connects to your Telegram account — you hand it an API hash, a phone number or a bot token, and it reads your dialogs, sends your messages, manages your groups. Useful, but it can only ever see what you already have access to.
This one is the opposite. It's a public catalog, so it answers the question none of the others can:
"What Telegram channels and groups exist about X, and which are worth joining?"
| account-based Telegram MCP | tgden MCP | |
|---|---|---|
| Credentials required | API ID/hash, phone or bot token | none |
| Install | local process (Python/Go/Node) | none — a hosted HTTP URL |
| Can see | chats you already joined | 1.4M+ public entities you haven't |
| Answers "what exists?" | no | yes |
| Can act on your account | yes | no — read-only by design |
Because it never touches an account, there's nothing to leak and nothing to get banned.
Quick start
Claude Code
claude mcp add --transport http tgden https://tgden.com/api/mcp
Claude Desktop · Cursor · any MCP client — add to your MCP config:
{
"mcpServers": {
"tgden": {
"type": "http",
"url": "https://tgden.com/api/mcp"
}
}
}
That's the entire setup: streamable HTTP transport, stateless, no auth handshake.
Check it by hand:
curl -s https://tgden.com/api/mcp \
-H 'Content-Type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | jq '.result.tools[].name'
Tools
| Tool | What it answers |
|---|---|
search_telegram |
Full-text search across channels, group chats and bots — with subscriber/member counts, categories and trust scores. |
best_channels_by_category |
"Best Telegram channels for crypto / AI / news / gaming…" — ranked by subscribers. |
channel_stats |
Everything known about one channel, by @username. |
search_posts |
Full-text search over recent posts — what channels actually said, not just their bio. |
market_listings |
Real classified listings (rent, sale, services, jobs) parsed from regional Telegram groups across 🇦🇪 🇹🇷 🇹🇭 🇻🇳 🇬🇪 🇺🇸. |
market_stats |
Listing totals by country and type. |
Resources
| URI | Contents |
|---|---|
tgden://catalog/top-channels |
Top-100 channels by subscribers. |
tgden://catalog/categories |
Full category taxonomy with channel counts. |
tgden://market/listings-latest |
The 100 freshest marketplace listings. |
tgden://market/stats |
Marketplace totals by country and type. |
Free REST API (no key either)
Telegram has no official public search. Finding a channel, a live discussion group or a bot by topic means guessing usernames or trusting closed directories. If you don't speak MCP, the same catalog is plain HTTP.
GET /api/catalog — search
| Param | Type | Default | Notes |
|---|---|---|---|
q |
string | — | Free-text query (title, username, topic) |
limit |
int | 20 |
Max results to return |
type |
string | all | channel · chat · bot (omit for everything) |
curl "https://tgden.com/api/catalog?q=crypto&type=channel&limit=3"
{
"items": [
{
"id": "6ec0407a-4f5e-4a86-9154-f42481bc9413",
"telegram_id": 2075341442,
"username": "hamster_kombat",
"title": "Hamster Kombat Announcement",
"is_private": false,
"avatar_url": "https://cdn4.telesco.pe/file/…"
}
]
}
GET /api/suggest — instant autocomplete
Grouped, ranked suggestions as-you-type. Good for search boxes and agent tool-use.
curl "https://tgden.com/api/suggest?q=btc"
{
"groups": [
{
"key": "chats",
"label": { "en": "live chats", "ru": "живые чаты" },
"items": [
{ "label": "BTC Times Discussion", "sub": "221 members",
"href": "/en/chat/thebtctimes", "kind": "chat", "username": "thebtctimes" }
]
}
]
}
Runnable examples: shell · Python · JavaScript.
Limits — published, because hidden limits are worse than low ones
| Surface | Limit |
|---|---|
REST (/api/catalog, /api/suggest) |
~200 requests / 10s per IP, shared fair-use |
MCP (/api/mcp) |
240 calls / hour per IP |
| Auth | none — no key, no signup, no email |
| Cost | free |
Verified search-engine and AI crawlers are excluded from the shared limit.
We treat these as a contract: when they change, we announce it here and in the docs. No silent throttling. Need guaranteed throughput or a bulk export for something real? Open an issue, or reach us via @tgden_bot.
About the data
Parsed continuously from Telegram's public surfaces — fresher than any LLM training set, which is most of the point of exposing it over MCP.
| Channels | 1,144,692 |
| Group chats | 181,772 |
| Bots | 142,084 |
| Indexed posts | 4,520,936 |
| Marketplace listings | 22,208 |
Census 31.07.2026. The MCP server reports these totals live, so what your agent sees is current — not whatever was true when this file was written.
Public data only: channels, groups and bots that are already publicly listed or linked. No private groups, no archives from closed communities, no personal data.
Use cases
- Agent tool-use — give an agent real Telegram discovery in one config line.
- Research — map communities, track niches, study the open Telegram graph.
- Discovery bots — recommend channels and chats by topic.
- Market intel — what's being rented, sold and offered in regional Telegram economies.
Machine-readable site summary for AI crawlers: https://tgden.com/llms.txt
License
MIT — do what you want, no warranty. Data is public Telegram metadata.
<sub>Maintained by tgden.com. Not affiliated with Telegram FZ-LLC.
Indexes public t.me links only.</sub>
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.