Anil-matcha/open-dots

Open-source, self-hosted AI agent workspace and alternative to OpenAI Dots, Meta Muse, Grok Bot, Instinct, Manus Cue, Claude Cowork, and ChatGPT agent....

View on GitHub ↗Jump to charts ↓Open shareable report →

Data as of . Signed-in members get hourly updates — create a free account.

Summary Information

Updated 12 minutes ago
Added to GitGenius on September 29th, 2026
Created on May 25th, 2023
Open Issues & Pull Requests: 3 (+0)
GitHub issues: Enabled
Number of forks: 665
Total Stargazers: 5,550 (+2)
Total Subscribers: 44 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 3.1 hours
Mean response time: 160.6 days
90th percentile: 882.7 days
Tracked items: 69

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 0
New in 7 days: 26
Closed in 7 days: 26
Avg open age: N/A days
Stale 30+ days: 0
Stale 90+ days: 0

Recent activity

Opened in 7 days: 1
Closed in 7 days: 1
Comments in 7 days: 0
Events in 7 days: 2

Top labels

No label distribution available yet.

Most active issues this week

Sign in to see which issues are moving.
Sign in

Detailed Description

Open Dots is a self-hosted AI workspace that provides an open-source alternative to OpenAI Dots, combining chat, tool use, approvals, connectors, and computer task capabilities in a local-first application.

The tool addresses the need for a transparent, inspectable AI workspace where users retain control over their data and explicitly approve higher-risk actions. It works by pairing a FastAPI backend with a Next.js frontend, storing conversations and state locally in SQLite while encrypting provider credentials at rest. Users create assistant personas with custom instructions and model IDs, stream chat responses with Markdown rendering and image attachment support, and route tool requests through a deny-by-default gateway that flags higher-risk actions for approval before execution. An optional Docker or Playwright-based computer runtime can be deployed locally or connected remotely to handle browser automation and system tasks.

Adoption suits developers and individuals who want to self-host an AI workspace with full visibility into operations and explicit control over approvals. It works best for local experimentation and single-user scenarios rather than multi-user hosting. The project is positioned as an alternative to OpenAI Dots for those prioritizing local data storage, configurable model access through a custom inference adapter, visible approval workflows, and an optional computer runtime under their control. The tool requires an external inference API that implements its specific request and response contract rather than the generic OpenAI-compatible chat completions interface.

The project is in active development as a prototype, with the maintainers explicitly noting that multi-user hosting and hostile-web isolation are not production ready. Development activity shows ongoing work across both backend and frontend components, with the codebase structured to support extensibility through connectors like Composio and configurable model providers. The tool remains under active iteration with known limitations acknowledged in provider and runtime documentation.