MCP Servers · groupultra

telegram-search

Telegram Search is a TypeScript application that exports and indexes Telegram chat history with fuzzy search and vector-based semantic search capabilities. It supports multi-language tokenization, image semantic search, and AI-powered summarization through local or cloud embeddings.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/groupultra/telegram-search

GitHub stars

261

Forks

TypeScript

Primary language

AGPL-3.0

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	groupultra/telegram-search
Owner	groupultra
Primary language	TypeScript
License	AGPL-3.0 — OSI-approved
Stars	4k
Forks	261
Open issues	53
Latest release	v1.2.8 (2026-06-29)
Last updated	2026-07-07
Source	https://github.com/groupultra/telegram-search

What telegram-search is

Built in TypeScript, the system uses PGlite or PostgreSQL with pgvector extension for semantic search, MinIO for media storage, and embedded LLM/vector models for message processing. It provides a web UI, Telegram Bot API integration, and real-time message synchronization with support for Chinese text tokenization and RAG-based question answering.

Quickstart

Get the telegram-search source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/groupultra/telegram-search.git2cd telegram-search3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Personal Telegram Archive & Knowledge Retrieval

Self-hosted backup of private Telegram conversations with semantic search, ideal for users managing large message volumes across multiple chats and needing rapid multi-language lookup without relying on Telegram's native search limitations.

Team/Community Message Management

Organizations running group chats or channels can deploy on-premises to index collective knowledge, enable unread message summaries, and provide context-aware AI assistant responses based on historical conversations.

AI-Enhanced Information Extraction

Leverage vector embeddings and RAG patterns to automatically summarize unread messages, build knowledge graphs from chat history, and provide intelligent Q&A over personal or team Telegram data.

Implementation considerations

Database choice: PGlite for single-user/small deployments, PostgreSQL with pgvector for multi-user or production. Verify pgvector availability and version compatibility.
Vector embeddings must be configured per-user in the settings UI (API keys for OpenAI, local models, etc.); plan how to manage API costs and fallback embeddings if external services are unavailable.
MinIO or local disk storage for media backups; ensure adequate capacity and configure retention/cleanup policies to avoid unbounded disk growth.
Telegram API credentials (API_ID, API_HASH) required from my.telegram.org; Bot Token optional for Telegram Bot integration. Rate limiting and account security are user's responsibility.
Multi-language tokenization is built-in (e.g., Chinese) but relies on internal library; test with your primary languages before production deployment.

When to avoid it — and what to weigh

Commercial SaaS Hosting Requirement — AGPL-3.0 license mandates source disclosure if modified and deployed as a service. Do not use as a commercial closed-source hosted platform without legal review and contributor agreement.
Minimal Infrastructure Preference — Deployment requires Docker, PostgreSQL/PGlite, MinIO (or local storage), and vector embeddings infrastructure. Not suitable for teams with zero DevOps capacity or preference for zero-config solutions.
Strict Data Residency or Compliance Isolation — The system auto-exports media to configured storage and may require external LLM/embedding APIs per user settings. If data must never leave a secure perimeter or requires HIPAA/FedRAMP certification, custom modifications are necessary.
High-Frequency Real-Time Sync at Scale — Architecture appears optimized for personal and small-team use; performance at thousands of concurrent users or millions of daily messages is unknown. Heavy real-time sync demands may require custom tuning.

License & commercial use

AGPL-3.0 (GNU Affero General Public License v3.0). This is a copyleft license requiring any modifications or network-accessible deployments to disclose full source code to users. Redistribution, modification, and private use are permitted under compliance.

Commercial use is legally possible under AGPL-3.0 but requires careful compliance: (1) if you modify the software, you must provide source access to users; (2) if you offer it as a service (SaaS), you must disclose source to your users; (3) you may charge for the service itself, but source must remain freely available. Legal review is strongly recommended before deploying as a commercial product. Requires review with your legal team.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Moderate
DEV.co fit	Good
Assessment confidence	High

Security considerations

User-supplied Telegram API credentials stored locally; ensure encryption at rest. Vector embeddings may be sent to external APIs (OpenAI, etc.) depending on user configuration; data residency and privacy are user's responsibility. Local message storage is plaintext unless encrypted at OS level. No explicit mention of input validation, rate limiting on search queries, or protection against prompt injection in RAG features. Review before exposing to untrusted users.

Alternatives to consider

Telegram Desktop native search

Built-in but lacks fuzzy and semantic search, Chinese tokenization, and AI summarization. Telegram Search adds these at the cost of self-hosting.

Memos or LogSeq (note-taking with search)

General-purpose knowledge bases with search; not Telegram-native. Better for multi-source data but requires manual capture of Telegram messages.

Custom Telegram bot + external database (DIY)

Full control and AGPL-free licensing, but requires significant engineering effort and ongoing maintenance versus using a pre-built, actively maintained solution.

Software development agency

Build on telegram-search with DEV.co software developers

Start with the free demo at search.lingogram.app, or deploy self-hosted with Docker Compose. Requires Telegram API credentials. Review AGPL-3.0 license for commercial use.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

telegram-mcp telegram-files typesense

Related on DEV.co

Explore the category and the services that help you build with it.

MCP Servers AI Development Web App Development API Development Talk to DEV.co

telegram-search FAQ

Can I use this for commercial SaaS hosting?

Not without modifications and legal review. AGPL-3.0 requires you to disclose source to users if you offer it as a service. You must either comply with AGPL (open-source your modifications) or obtain alternative licensing from the authors.

What are the hardware requirements?

For small deployments: 4GB RAM, 20GB disk (user messages + embeddings + media). PostgreSQL + pgvector and vector indexing scale with message volume; no published benchmarks provided. Requires review for your scale.

Does Telegram Search store my messages on external servers?

By default, messages are stored in your local PGlite or PostgreSQL. Media files go to MinIO (local or cloud, your choice). LLM/embedding calls depend on your per-user API key configuration; if you use OpenAI or cloud services, those calls send embeddings externally. No data is sent to Telegram Search creators' servers unless you use the hosted demo at search.lingogram.app.

Is Chinese/non-English search supported?

Yes; multi-language tokenization is built-in, with explicit mention of Chinese support. Test with your specific language and variant before production.

Software development & web development with DEV.co

DEV.co is a software development agency delivering custom software development services to companies building on open source. Our software developers and web developers design, integrate, and ship production systems — spanning web development, APIs, AI, data, and cloud. If telegram-search is part of your mcp servers roadmap, our team can implement, customize, migrate, and maintain it.

Ready to Index Your Telegram History?

Start with the free demo at search.lingogram.app, or deploy self-hosted with Docker Compose. Requires Telegram API credentials. Review AGPL-3.0 license for commercial use.

Talk to DEV.co Schedule a Consultation