traul

Index Slack, Telegram, Discord, Linear, Gmail, and WhatsApp messages locally for unified search.

111|9|Updated Mar 14, 2026

Search Tags:#embeddings #search #channels #sqlite #signals #briefing #traul

Ready to run

npx skills add https://github.com/dandaka/traul --skill traul

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: traul
Source: https://github.com/dandaka/traul/tree/main
Command: npx skills add https://github.com/dandaka/traul --skill traul

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires telethon, PyYAML, and includes scripts (resource) components.

What problem does it solve?

Traul indexes and makes searchable your multi-source conversations locally, enabling fast recall, cross-channel search, and proactive signals without leaving your device.

Core Features & Use Cases

Unified local index of Slack, Telegram, Discord, Linear, Gmail, Claude Code sessions, Markdown files, and WhatsApp for fast search and recall.
Hybrid vector + keyword search with optional embeddings, plus a signals/briefing workflow to surface proactive insights.
Browse channels, get full threads, and run ad-hoc SQL-like queries against the data locally.

Quick Start

Sync data with Traul by running traul sync, then perform a sample search such as traul search 'deployment'.

Frequently Asked Questions about traul

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I search across Slack, Telegram, and Discord conversations locally?▼

You can search across Slack, Telegram, and Discord conversations locally by syncing them into a unified local index. This enables fast cross-channel recall and hybrid vector + keyword search directly on your device.

What is the best way to index WhatsApp and Gmail messages for offline search?▼

Indexing WhatsApp and Gmail messages for offline search is best handled by a local-first memory tool that uses SQLite with FTS5 for text search. This setup allows you to run ad-hoc SQL-like queries and retrieve full threads without relying on cloud services.

Can I use SQLite and sqlite-vec for hybrid vector and keyword search?▼

Yes, you can use SQLite and sqlite-vec for hybrid vector and keyword search. This architecture uses SQLite as the core database with FTS5 for text search and sqlite-vec for embeddings, optionally leveraging Ollama for local embeddings.

How do I set up proactive signals and briefings from multiple messaging sources?▼

To set up proactive signals and briefings from multiple messaging sources, you need a signals workflow that tracks and indexes sources like Linear and Claude Code sessions locally. This surfaces proactive insights and filters by channel and source automatically.

Does local conversation search work with Claude Code sessions and Markdown files?▼

Local conversation search does work with Claude Code sessions and Markdown files. The system indexes these sources alongside communication platforms, allowing you to browse channels, get full threads, and perform unified search.