Screenpipe AI Memory

Search screenpipe's recorded content: screen text (accessibility APIs, with OCR fallback), audio transcriptions, and UI elements. Returns timestamped results with app context. Call with no parameters to get recent activity. Use the 'screenpipe://context' resource for current time when building time-based queries. WHEN TO USE WHICH content_type: - For meetings/calls/conversations: content_type='audio', do NOT use q param (transcriptions are noisy, q filters too aggressively) - For screen text/reading: content_type='all' or 'accessibility' - For time spent/app usage questions: use activity-summary tool instead (this tool returns content, not time stats) SEARCH STRATEGY: First search with ONLY time params (start_time/end_time) — no q, no app_name, no content_type. This gives ground truth of what's recorded. Scan results to find correct app_name values, then narrow with filters using exact observed values. App names are case-sensitive (e.g. 'Discord' vs 'Discord.exe'). The q param searches captured text, NOT app names. NEVER report 'no data' after one filtered search — verify with unfiltered time-only search first. DEEP LINKS: When referencing specific moments, create clickable links using IDs from search results: - OCR results (PREFERRED): [10:30 AM — Chrome](screenpipe://frame/12345) — use content.frame_id from the result - Audio results: [meeting at 3pm](screenpipe://timeline?timestamp=2024-01-15T15:00:00Z) — use exact timestamp from result NEVER fabricate frame IDs or timestamps — only use values from actual search results.

Export a video of screen recordings for a specific time range. Creates an MP4 video from the recorded frames between the start and end times. IMPORTANT: Use ISO 8601 UTC timestamps (e.g., 2024-01-15T10:00:00Z) or relative times (e.g., '16h ago', 'now') EXAMPLES: - Last 30 minutes: Calculate timestamps from current time - Specific meeting: Use the meeting's start and end times in UTC

List detected meetings with duration, app, and attendees. Returns meetings detected via app focus (Zoom, Meet, Teams) and audio. Only available when screenpipe runs in smart transcription mode.

Get a lightweight compressed activity overview for a time range (~200-500 tokens). Returns app usage (name, frame count, active minutes, first/last seen), recent accessibility texts, and audio speaker summary. Minutes are based on active session time (consecutive frames with gaps < 5min count as active). first_seen/last_seen show the wall-clock span per app. USE THIS TOOL (not search-content or raw SQL) for: - 'how long did I spend on X?' → active_minutes per app - 'which apps did I use today?' → app list sorted by active_minutes - 'what was I doing?' → broad overview before drilling deeper - Any time-spent or app-usage question WARNING: Do NOT estimate time from raw frame counts or SQL queries — those are inaccurate. This endpoint calculates actual active session time correctly.

Search structured UI elements (accessibility tree nodes and OCR text blocks). Returns ~100-500 bytes per element — much lighter than search-content for targeted lookups. Each element has: id, frame_id, source (accessibility/ocr), role (AXButton, AXStaticText, AXLink, etc.), text, bounds, depth. Use for: finding specific buttons, links, text fields, or UI components. Prefer this over search-content when you need structural UI detail rather than full screen text.

Search recent screen activity

Find content from a specific application

Get audio transcriptions from meetings

Create a new screenpipe pipe (scheduled AI automation)

Current date/time and pre-computed timestamps for common time ranges

How to use screenpipe search effectively

Interactive search UI for exploring screen recordings and audio transcriptions

How to create screenpipe pipes (scheduled AI automations): format, YAML frontmatter, schedule syntax, API parameters, and example templates

Full screenpipe REST API reference: search, activity-summary, elements, frames, export, retranscribe, raw SQL, connections, speakers (60+ endpoints)