Frequently Asked Questions

Common questions about screenpipe pricing, privacy, storage, and platform support

Frequently Asked Questions

Answers to the most common questions about screenpipe.

General

What is screenpipe?

screenpipe is an open source application that continuously captures your screen and audio, creating a searchable, AI-powered memory of everything you do on your computer. All data is stored locally on your device by default.

Think of it as giving AI perfect memory of everything you’ve seen, heard, or done on your computer - so you can search, recall, and automate based on your actual work history.

Is screenpipe free?

The core engine is open source (MIT license) and free to inspect and modify. The desktop app requires a one-time lifetime purchase of $400.

There are no recurring subscription costs required for core features. Optional Pro subscription ($39/month) adds cloud sync and priority support.

How is this different from Rewind/Limitless or Microsoft Recall?

Key differences:

• Open source: Full transparency, no vendor lock-in
• Local-first: Your data never leaves your device by default
• Multi-monitor: Captures all monitors (not just active window)
• Developer API: Full REST API and SDK for custom integrations
• Plugin system: AI agents (Pipes) for automation
• Cross-platform: Works on macOS, Windows, and Linux

See our detailed comparisons:

• vs Rewind/Limitless
• vs Microsoft Recall

Privacy & Security

Does screenpipe send my data to the cloud?

No. By default, all data is stored locally on your device in a SQLite database. Nothing is sent to external servers.

You can optionally enable:

• Cloud AI models (OpenAI, Anthropic) for better search
• Encrypted sync between your devices (zero-knowledge encryption)

But the core app works 100% offline with no cloud dependency.

Can I use screenpipe with fully local AI?

Yes! screenpipe has built-in support for Ollama, allowing you to run AI models completely locally:

# Install Ollama
curl -fsSL https://ollama.ai/install.sh | sh

# Pull a model
ollama pull llama2

Then configure screenpipe to use your local Ollama endpoint. No data leaves your device.

Is the source code auditable?

Yes. screenpipe is 100% open source under the MIT license. You can:

• View all code on GitHub
• Audit security and privacy practices
• Build from source
• Contribute improvements
• Fork for custom needs

What about sensitive data like passwords?

screenpipe captures whatever is visible on your screen, so it’s important to configure exclusions:

• Exclude apps: Block password managers, banking apps, etc.
• Exclude windows: Filter by window title patterns
• Exclude URLs: Skip specific websites
• Time restrictions: Only capture during work hours

All these settings are configurable in the app settings.

Can I delete my data?

Yes, you have full control:

• Delete specific time ranges via the UI
• Clear all data from settings
• Delete the SQLite database file directly
• Use the API to programmatically manage data

Your data is stored in ~/.screenpipe/ (or Windows equivalent).

Storage & Performance

How much disk space does screenpipe use?

Typically 5-10 GB per month.

screenpipe uses event-driven capture - it only saves frames when something changes (app switch, click, typing, scroll). This reduces storage by 70-80% compared to continuous recording.

Actual usage depends on:

• Number of monitors
• How actively you use your computer
• Whether you enable audio recording
• Capture quality settings

Does screenpipe slow down my computer?

No. Typical resource usage:

• CPU: 5-10% on modern hardware
• RAM: 0.5-3 GB
• Disk: Minimal writes (event-driven)

Event-driven capture means screenpipe only processes frames when something actually changes, keeping CPU usage low.

What are the minimum system requirements?

• RAM: 8 GB recommended (4 GB minimum)
• Storage: ~10 GB free space to start
• CPU: Any modern processor (2015+)
• OS: macOS 10.15+, Windows 10+, or Linux

screenpipe works on modest hardware thanks to event-driven capture.

How long can I keep data?

As long as you have disk space. screenpipe doesn’t auto-delete old data.

You can:

• Keep years of history if you have storage
• Set up automatic cleanup via Pipes
• Manually archive or delete old data
• Store data on external drives

Platform Support

What platforms are supported?

Platform	Support	Installation
macOS (Apple Silicon)	✅ Full	Native .dmg installer
macOS (Intel)	✅ Full	Native .dmg installer
Windows 10/11	✅ Full	Native .exe installer
Linux	✅ Supported	Build from source

Can I use screenpipe on multiple computers?

Yes! Options:

1. Install separately: Each device has its own local data
2. Pro sync: Cloud sync between devices (encrypted, $39/month)
3. Manual export: Export/import data via API

With Pro sync, you can search across all your devices from any device.

Does it work with multiple monitors?

Yes! screenpipe captures all connected monitors simultaneously.

This is a key advantage over Rewind/Limitless (which only captures the active window).

Can I use it on Linux?

Yes. Linux support is available by building from source:

git clone https://github.com/screenpipe/screenpipe
cd screenpipe
cargo build --release

Linux support uses Tesseract for OCR when accessibility data isn’t available.

Features

Does screenpipe record audio?

Yes. screenpipe can record:

• System audio: What you hear (meetings, videos, music)
• Microphone: What you say

Audio is transcribed locally using OpenAI Whisper, making it fully searchable.

You can enable/disable audio recording independently in settings.

How does text extraction work?

screenpipe uses a hybrid approach:

1. Accessibility tree (primary): Extracts structured text from the OS (buttons, labels, text fields) - fast and accurate
2. OCR fallback: When accessibility isn’t available (remote desktops, games), uses:
   • Apple Vision on macOS
   • Windows native OCR
   • Tesseract on Linux

This gives you high accuracy with minimal CPU usage.

Can I search with natural language?

Yes! screenpipe supports:

• Keyword search: Fast full-text search
• Semantic search: Natural language queries using AI embeddings
• Filters: By app, window title, URL, date range, content type

Example: “Find that email about the Q4 budget from last week”

What are Pipes?

Pipes are scheduled AI agents defined as markdown files. Each pipe:

• Contains a prompt and schedule
• Can query your screen/audio data
• Can call APIs and write files
• Runs automatically in the background

Built-in pipes:

• Obsidian sync (auto journal)
• Reminders (extract todos)
• Idea tracker (surface insights)

Create custom pipes by writing markdown in ~/.screenpipe/pipes/.

Can I integrate with ChatGPT/Claude/Cursor?

Yes! screenpipe runs as an MCP server:

# Add to Claude Desktop
claude mcp add screenpipe -- npx -y screenpipe-mcp

This allows AI assistants to:

• Search your screen history
• Get recent context
• Access meeting transcriptions
• Build context-aware responses

Works with: Claude Desktop, Cursor, VS Code (Cline, Continue), and any MCP-compatible client.

Pricing

What does the lifetime license include?

The $400 lifetime purchase includes:

• Desktop app for all platforms
• All current features
• All future updates forever
• Multi-monitor support
• Audio transcription
• Full API access
• Plugin system (Pipes)
• Local AI support

No recurring subscription required.

What's included in Pro?

Pro subscription ($39/month) adds:

• Cloud sync: Sync data between devices (encrypted)
• Priority support: Faster response times
• Pro AI models: Access to premium AI capabilities

Core features work without Pro - it’s entirely optional.

Is there a free trial?

You can try the open source CLI version for free:

npx screenpipe@latest record

This gives you the core capture and search functionality to evaluate before purchasing.

What about team/enterprise pricing?

screenpipe Teams provides:

• Central config management
• Shared AI pipes
• Per-pipe data permissions
• Admin dashboard
• MDM deployment (Intune, SCCM)
• SSO/SAML (Enterprise)
• Audit logs (Enterprise)

Pricing is custom based on team size. Contact sales for details.

Teams & Enterprise

Can I deploy screenpipe to my team?

Yes. screenpipe Teams lets you:

• Deploy capture settings to all employees
• Share AI workflows (Pipes) team-wide
• Control what AI can access with per-pipe permissions
• Employees’ data stays on their devices
• Admin dashboard for management

See screenpi.pe/team for details.

How do AI data permissions work?

Each pipe (AI agent) can have YAML frontmatter controlling access:

allow-apps: ["Slack", "Zoom"]
deny-windows: ["*password*", "*bank*"]
allow-content-types: ["ocr", "audio"]
time-range: "09:00-18:00"
days: "Mon,Tue,Wed,Thu,Fri"
allow-raw-sql: false

Enforced at three OS-level layers - not by prompting the AI. Even compromised agents can’t access denied data.

Is screenpipe compliant with security standards?

screenpipe Teams Enterprise offers:

• SOC 2 ready: Compliance framework support
• HIPAA ready: Healthcare data protection
• Audit logs: Complete activity tracking
• SSO/SAML: Enterprise authentication
• MDM deployment: Intune, SCCM, Robopack

Contact sales for compliance documentation.

Troubleshooting

Where can I get help?

• Documentation: docs.screenpi.pe
• Discord: discord.gg/screenpipe
• GitHub: Issues
• Email: [e-mail]

For common issues, see our troubleshooting guide.

How do I report a bug?

1. Check existing issues
2. Open a new issue with:
   • OS and version
   • screenpipe version
   • Steps to reproduce
   • Logs from ~/.screenpipe/logs/
3. Join Discord for real-time help

---

Still have questions?

Join our Discord community or check the troubleshooting guide.