@screenpipe/sync

Sync your Screenpipe data to remote AI agents.

One-liner to sync daily summaries to a remote server (e.g., OpenClaw). Uses Claude Code CLI for AI-powered extraction.

Quick Start

bash

# AI summary to stdout
bunx @screenpipe/sync

# Save daily summaries locally
bunx @screenpipe/sync --output ~/Documents/brain/context --git

# Sync summaries to remote server
bunx @screenpipe/sync --output /tmp/summaries --remote user@host:~/screenpipe-pkm

# Persistent daemon (survives reboot)
bunx @screenpipe/sync --daemon --output /tmp/summaries --remote user@host:~/clawd/screenpipe-pkm

What It Extracts

Category	Description
Todos	Action items visible on screen or mentioned
Goals	Objectives, intentions, targets mentioned
Decisions	Choices made or discussed
Activities	Key tasks worked on, by app
Meetings	Calls, conversations, collaborations
Blockers	Problems, frustrations, obstacles
Insights	AI observations about work patterns

Example Output

markdown

# Daily Context - 2026-01-31

> Analyzed 480 minutes of screen activity

## Apps Used
- **VS Code**: ~180 min
- **Chrome**: ~120 min
- **Slack**: ~60 min

## Todos Extracted
- Fix authentication bug in login.ts
- Review PR #234 for payment integration
- Send weekly update to investors

## Goals Mentioned
- Ship v2.9 by Friday
- Reach 50% activation rate

## AI Insights
- Heavy context switching between Slack and VS Code
- Deep focus block from 2-4pm on auth refactor

Options

Flag	Description	Default
`-o, --output <dir>`	Save to directory	stdout
`-h, --hours <n>`	Hours to analyze	12
`-g, --git`	Auto commit & push	false
`-r, --remote <host>`	Sync via SSH	-
`-d, --daemon`	Install persistent sync	false
`--interval <secs>`	Daemon interval	3600
`--stop`	Stop daemon	-
`--json`	JSON output	markdown
`-v, --verbose`	Debug output	false

AI Summarization

Uses Claude Code CLI if available:

bash

# Install Claude Code CLI
npm install -g @anthropic-ai/claude-code

# Verify it works
claude --version

Falls back to structured extraction (no AI) if CLI not found.

OpenClaw Integration

For daily digests via Telegram:

Local Mac syncs summaries hourly:

bash

bunx @screenpipe/sync --daemon --output /tmp/summaries --remote openclaw:~/clawd/screenpipe-pkm

OpenClaw has native cron that reads files and sends digest:

bash

# Already configured - runs at 9pm PT
sudo docker exec openclaw-gateway node dist/index.js cron list

Use Cases

Daily Journaling

bash

bunx @screenpipe/sync --output ~/journal --hours 16

Weekly Review Prep

bash

bunx @screenpipe/sync --hours 168 --json > week.json

Daemon Mode (Recommended)

bash

# Install persistent sync - survives reboot
bunx @screenpipe/sync --daemon -r user@host:~/screenpipe-pkm

# Check status
# macOS: cat /tmp/screenpipe-sync.log
# Linux: systemctl --user status screenpipe-sync.timer

# Stop daemon
bunx @screenpipe/sync --stop

How It Works

Query - Fetches OCR data from local Screenpipe API
Dedupe - Removes duplicate/similar screen captures
Extract - Claude Code CLI analyzes for structured data
Format - Outputs markdown or JSON
Sync - SCPs to remote server

Requirements

Screenpipe running locally
Bun runtime
Claude Code CLI (optional, for AI extraction)

Privacy

All processing happens locally. Screen data never leaves your machine unless you explicitly sync summaries to a remote.

License

MIT - Part of the Screenpipe project.