Skip to content
Rawdash

Introduction

What is Rawdash?

Rawdash is an open-source headless dashboard backend — a TypeScript library you embed in your existing server to handle the data side of dashboards as a solved problem.

It provides:

  • Pre-built connectors for SaaS tools (GitHub, Slack, Linear, and more)
  • A background sync engine that keeps data fresh automatically
  • A pluggable storage layer (ships with a Turso/libSQL adapter)
  • A clean HTTP API any frontend can query

You bring the UI. Rawdash handles the plumbing.

Architecture overview

┌─────────────────────────────────────────────┐
│                 Your server                  │
│                                              │
│  rawdash.config.ts                           │
│  ├── connectors  ──► background sync         │
│  ├── storage     ──► Turso / custom adapter  │
│  └── dashboards  ──► widget definitions      │
│                                              │
│  Mount: app.use('/api/rawdash', handler)     │
└──────────────────┬──────────────────────────┘
                   │  REST / MCP
        ┌──────────▼──────────┐
        │   Your frontend /   │
        │   Claude / any client│
        └─────────────────────┘

When to use Rawdash

Rawdash is a good fit when you:

  • Need a dashboard that aggregates data from multiple SaaS tools
  • Want synced, cached data rather than live API calls on every page load
  • Prefer a TypeScript-first, code-defined config over a YAML/JSON schema
  • Want to self-host on your own infra

Core concepts

Connectors

A connector wraps a third-party API. It knows how to fetch data, what shape to store it in, and how often to refresh. You configure connectors in your defineConfig call.

Dashboards

A dashboard groups one or more connectors under a single ID. Your frontend queries dashboards by ID.

Storage adapter

The storage adapter persists synced data. The built-in adapter uses Turso (libSQL), but the interface is small enough to implement for any database.

Retention

Each connector can declare a retention policy — how long to keep historical snapshots before pruning old rows.

Next steps