OpenWorker
Source Download →

AI that gets your everyday tasks done.

Ask for an outcome, not just an answer. OpenWorker works in the tools you use every day — your files, Slack, calendar, and more — and carries tasks through from start to finish: delivering a polished document, sending a Slack message, or updating a calendar entry. It checks in before important actions, so you stay in charge.

100% open source Bring your own model and API keys Local-first and privacy-preserving
Works with what you choose

Your models. Your tools.

Use closed-weight models, open-weight models, or run locally. Connect the everyday tools where your work already happens.

Model providersCloud or fully local
OpenAI Anthropic Gemini Z AI · GLM DeepSeek Kimi MiniMax Qwen xAI Mistral Together AI Fireworks Ollama
Everyday toolsUse one-click or manual connections
Slack Outlook Gmail Google Calendar Notion HubSpot GitHub Attio Google Drive Jira Linear Asana Dropbox Box Figma GitLab More available
Out of the box

Ready for real work, not just chat.

OpenWorker comes ready for common kinds of work — with the right tools, working style, and check-ins already set up. Choose a use case and hand off a task.

Sales

Researches accounts, preps every meeting, drafts follow-ups that sound like you.

Works with — HubSpot · Email · Calendar
Checks in — before anything is sent

Executive Assistant

Owns the calendar, triages the inbox, and keeps your day on rails.

Works with — Email · Calendar · Slack
Checks in — before replying or moving meetings

Marketing

Tracks what's working — spend, attribution, and the Monday report, without five dashboards.

Works with — HubSpot · GA4 (soon) · Slack
Checks in — before anything is published

Ops On-call

Triages alerts, digs through runbooks, drafts the incident notes nobody wants to write.

Works with — Slack · PagerDuty (soon) · GitHub (soon)
Checks in — before touching anything live
Works in the background

Starts when work happens.

Set a trigger once. OpenWorker starts the task and checks in before important actions.

Slack mention Triage the alert

Check the runbook and draft an in-thread response.

Incoming email Prepare the reply

Gather the answers and hold the draft for review.

Your schedule Deliver the report

Pull the numbers and finish it before standup.

How it works

Up and running in three steps.

1

Bring your model

Paste an API key for OpenAI, Anthropic, Google, or a supported provider offering open-weight models such as Kimi, GLM, or DeepSeek. Your key is stored on your machine. Or choose Ollama to run a local model with no API key.

~30 seconds
2

Connect your tools

Connect Slack, email, calendar, HubSpot, Notion, and more. Use one-click sign-in where available, or paste your own tokens. Tokens are stored on your machine.

One-click sign-in, or your own tokens
3

Get it done

Ask for what you need — a brief, a report, a tidy inbox. OpenWorker does the work, checks in when it matters, and hands you the deliverable.

You stay in charge
FAQ

Questions, answered.

Is it really free? +

Yes — the app is open source. Bring your own API key for OpenAI, Anthropic, Google, or another supported provider, or run a local model with Ollama. You pay your model provider directly and nothing to us.

Where do my passwords and tokens live? +

On your device, in the app's local secret store. The optional cloud account only brokers the OAuth handshake — tokens are handed straight to your machine and never stored in the cloud.

Will it do things without asking me? +

Only what you've allowed. Anything consequential — sending an email, posting to Slack, running a command — lands as a check-in you approve, in the app or from Slack. You choose how independently it’s allowed to act.

Do I need an account? +

No. Everything works while signed out, using API keys and tokens you provide. A free OpenWorker account adds one-click connections to tools such as Slack, where vendor restrictions require OpenWorker to manage the sign-in flow.

Which models can I use? +

OpenAI, Anthropic, and Google models through your own API keys; open-weight models such as Kimi, GLM, and DeepSeek through supported providers; or local models through Ollama. You can switch models per task, even mid-conversation.

Ready when you are

Your first deliverable is a
download away.

Free, open source, and on your machine in under a minute.

Download for Mac — free → View on GitHub