ChatGPT — use your subscription with screenpipe

Connect your ChatGPT Plus or Pro subscription to power screenpipe’s AI chat and pipes — no API key needed.

if you have a ChatGPT Plus or Pro subscription, you can connect it directly to screenpipe. this lets you use OpenAI’s latest models for chat, summaries, and pipes without managing API keys or paying extra.

what you get

• AI chat — ask questions about your screen history, meetings, and activity using GPT models
• pipes — run scheduled or on-demand automations (summaries, time tracking, etc.) powered by your subscription
• no API key needed — sign in with your OpenAI account, that’s it
• use your existing subscription — no extra costs beyond your ChatGPT Plus/Pro plan

how to connect

1. Open model selector

   open screenpipe and click the model selector next to the chat input (shows “OAI”, “Ollama”, etc.)

2. Select ChatGPT

   select ChatGPT as your provider from the dropdown

3. Sign in

   click sign in with ChatGPT — a browser window opens

4. Authorize access

   log in with your OpenAI account and approve access to screenpipe

5. Done

   you’ll see a green checkmark and your available models. start chatting!

your ChatGPT credentials are stored locally on your machine. screenpipe never sees your password.

available models

after signing in, you can pick from your subscription’s available models:

subscription	models available
ChatGPT Plus	GPT-4o, GPT-4o mini, and more
ChatGPT Pro	GPT-4o, GPT-4o mini, o1, o1-pro, and more

the model list updates automatically based on what your subscription includes. pick the one that fits your needs:

• GPT-4o mini - fastest, great for quick queries and summaries
• GPT-4o - balanced speed and capability
• o1 - advanced reasoning (Pro only)
• o1-pro - maximum reasoning capability (Pro only)

using ChatGPT in screenpipe chat

once connected, use ChatGPT models in screenpipe’s AI chat:

> what did I work on this morning?

> summarize my meeting from 2pm today

> find that documentation I was reading about React hooks

> which apps did I use most yesterday?

screenpipe automatically queries your screen and audio history and sends context to ChatGPT.

using ChatGPT for pipes

pipes are automations that run on a schedule or on demand. when you connect your ChatGPT subscription, pipes can use it automatically.

setting up a pipe with ChatGPT

1. Go to pipes

   open pipes in the screenpipe sidebar

2. Select or create a pipe

   pick an existing pipe (like “toggl sync” or “day recap”) or create your own

3. Configure AI preset

   in the pipe settings, select your ChatGPT preset as the AI provider

4. Enable the pipe

   toggle the pipe on to start using it

example pipes

toggl sync - auto-track time in Toggl based on your screen activity

• runs every 30 minutes
• analyzes what apps you used
• creates time entries automatically

day recap - generate a summary of your day

• one-click to generate
• summarizes meetings, tasks, and screen activity
• perfect for daily standup prep

meeting summary - summarize meeting transcripts

• processes audio transcriptions
• extracts action items and key points
• can post to Slack or Notion

how it’s different from MCP

screenpipe also supports MCP integrations with ChatGPT Desktop and other apps. here’s the difference:

	ChatGPT subscription	MCP integration
what it does	powers screenpipe’s built-in AI chat and pipes	lets external apps query screenpipe data
setup	sign in with OpenAI account inside screenpipe	add screenpipe as MCP server in external app
where you chat	in screenpipe’s chat window	in the external app (ChatGPT Desktop, etc.)
runs pipes	yes	no
use case	using screenpipe as your AI interface	using external apps with screenpipe data

you can use both at the same time.

ChatGPT Desktop via MCP

if you want to use ChatGPT Desktop (the native app) with screenpipe:

1. ChatGPT Desktop doesn’t support MCP yet (as of March 2026)
2. use Claude Desktop or Cursor for MCP-based integrations
3. or use screenpipe’s built-in chat with your ChatGPT subscription (this page)

OpenAI hasn’t announced MCP support for ChatGPT Desktop yet. check MCP server docs for updates.

signing out

to disconnect your ChatGPT account:

1. Open model selector

   click the model selector in screenpipe

2. Sign out

   click the sign out button next to your ChatGPT connection

this removes the stored tokens from your machine. your screen data stays local.

troubleshooting

sign-in window doesn’t open?

• make sure screenpipe is running
• check your default browser is set
• try restarting the app and signing in again

“could not get token” error?

• your session may have expired — sign out and sign back in
• check that your ChatGPT subscription is active at openai.com
• ensure you’re not blocking pop-ups in your browser

models not showing up?

• sign out and sign back in to refresh your token
• some models are only available on higher subscription tiers
• verify your subscription at openai.com/account

pipes failing with auth errors?

• tokens refresh automatically, but if it persists, sign out and back in
• check screenpipe logs for specific error messages
• ensure your OpenAI account is in good standing

rate limit errors?

• you’re hitting your subscription’s rate limits
• try using a faster model (GPT-4o mini)
• reduce pipe frequency if they’re running too often

chat responses are slow?

• use GPT-4o mini for faster responses
• reduce the time range you’re querying
• avoid including screenshots in queries unless needed

still stuck? ask in our discord — we can help debug authentication issues.

privacy & security

• your screen data is only sent to OpenAI when you explicitly chat or run a pipe
• all authentication tokens are stored locally on your machine
• screenpipe never sees your OpenAI password
• you can revoke access anytime from your OpenAI account settings
• screen data is sent directly from your machine to OpenAI (not through screenpipe servers)

resources

• OpenAI ChatGPT
• screenpipe pipes documentation
• API reference