Create a secure chatbot that knows everything about what you’ve seen, said, or heard
I’ve built a tool that record your screen and microphone 24/7, while keeping the data 100% on your computer:
https://github.com/louis030195/screen-pipe
AI is very powerful nowadays but barely know anything about you or your company in particular.
Screenpipe is a library that allows you to gather all your life context and connect it to LLMs easily for:
- search (e.g. go beyond your limited human memory)
- automation (such as making actions on the web while you work, syncing company’s knowledge, etc.)
- etc.
Get started
You need to install a few dependencies:
# On Mac
brew install ffmpeg
# On Linux
sudo update
sudo apt install -y ffmpeg libasound2-dev libavcodec-dev libavformat-dev libavutil-dev
Also install Rust.
Clone screenpipe:
git clone https://github.com/louis030195/screen-pipe
cd screen-pipe
Build screenpipe:
# This runs a local SQLite DB + an API + screenshot, ocr, mic, stt, mp4 encoding
# if you are on mac, use "cargo build --release --features metal" to leverage M series acceleration
cargo build --release
./target/release/pipe
# or only stream audio + speech to text to stdout
./target/release/pipe-audio
# or only stream screenshots + ocr to stdout
./target/release/pipe-vision
# or only record mp4 videos + json containing ocr
./target/release/pipe-video
Update the variables in this directory (cp .env.example .env) :
# You must first activate a Billing Account here: https://platform.openai.com/account/billing/overview
# Then get your OpenAI API Key here: https://platform.openai.com/account/api-keys
OPENAI_API_KEY=XXXXXXXX
# Generate a random secret: https://generate-secret.vercel.app/32 or `openssl rand -base64 32`
AUTH_SECRET=XXXXXXXX
# Instructions to create kv database here: https://vercel.com/docs/storage/vercel-kv/quickstart and
KV_URL=XXXXXXXX
KV_REST_API_URL=XXXXXXXX
KV_REST_API_TOKEN=XXXXXXXX
KV_REST_API_READ_ONLY_TOKEN=XXXXXXXX
And run the web app:
cd examples/ts/vercel-ai-chatbot
pnpm i # or npm i
pnpm dev # or npm run dev
Enjoy!
If you like screenpipe, please drop a star:
https://github.com/louis030195/screen-pipe
Struggle to get it running? I’ll install it with you in a 15 min call.
Louis