FAQ

<Expandable title="what are pipes?"> pipes are scheduled AI agents that run on your screen data. each pipe is a `.md` file with a prompt and a schedule. an AI agent reads the prompt, queries the screenpipe API, and takes action.

examples: sync to Obsidian, track time in Toggl, send daily summaries.

see the pipes guide for details. </Expandable>

<Expandable title="what are the hardware requirements?"> - **minimum**: dual-core CPU, 2GB RAM, 20GB disk - **recommended**: quad-core, 4GB+ RAM, 50GB+ SSD - runs 24/7 on a MacBook Pro M3 or a $400 Windows laptop - ~600 MB RAM, ~10% CPU, ~30 GB/month storage at 1 FPS </Expandable> <Expandable title="where is data stored?"> all data is in `~/.screenpipe/`: - `db.sqlite` — metadata, OCR text, transcriptions - `data/` — MP4 screen recordings, audio chunks - `pipes/` — installed pipes

to backup, copy ~/.screenpipe/ to another location. </Expandable>

<Expandable title="how do i contribute?"> - check [github issues](https://github.com/screenpipe/screenpipe/issues) - join [discord](https://discord.gg/screenpipe) - follow [contribution guidelines](https://github.com/screenpipe/screenpipe/blob/main/CONTRIBUTING.md) </Expandable> <Expandable title="what OCR engines are supported?"> `apple-native` (macOS), `windows-native` (Windows), `tesseract` (Linux), `unstructured` (cloud), `custom` </Expandable> <Expandable title="is my data private?"> yes. everything runs locally. no data leaves your machine unless you explicitly choose a cloud provider (deepgram, unstructured). you control what's captured with `--ignored-windows` and `--included-windows`. </Expandable> <Expandable title="how do i connect screenpipe to AI?"> three ways: 1. **MCP server** — works with Claude Desktop, Cursor ([guide](/mcp-server)) 2. **pipes** — scheduled AI agents ([guide](/pipes)) 3. **REST API** — `curl http://localhost:3030/search?q=your+query` </Expandable> <Expandable title="music sounds bad or keeps pausing with bluetooth headphones"> this is a macOS bluetooth limitation, not a screenpipe bug. when any app opens a bluetooth microphone, macOS switches the headset from the high-quality A2DP codec (stereo, full bass) to the low-quality HFP codec (phone quality, 8kHz mono). this affects all audio — music, ANC, everything. the same thing happens with Zoom, Google Meet, or any app that uses your bluetooth mic.

fix: go to screenpipe settings → audio devices, and switch the input device from your bluetooth headset to your MacBook's built-in microphone. screenpipe will stop opening the bluetooth mic, macOS stays on the high-quality codec, and your music sounds normal again. screenpipe still captures your voice through the built-in mic.

note: AirPods don't have this issue — Apple uses a proprietary codec that handles input and output simultaneously without switching profiles. </Expandable>

<Expandable title="what speech-to-text engines are supported?"> **local (default)**: `parakeet` (25 languages, fastest), `whisper-large-v3-turbo` (99 languages), `whisper-tiny` (lightweight)

cloud: deepgram / screenpipe-cloud (highest accuracy)

on macOS Apple Silicon, parakeet uses Metal GPU acceleration via MLX for 8x faster transcription. </Expandable>