
Screenpipe Api
Query screen recordings, audio, and UI data via local Screenpipe REST API
What You Can Do
Query your local Screenpipe instance to retrieve screen recordings, audio transcriptions, UI element accessibility trees, keyboard/mouse input logs, and productivity analytics. You can search by keywords, filter by content type (audio, OCR, accessibility, input), set time ranges, and extract structured data about your applications, meetings, and work sessions without sending data to external servers.
Features
query screen activity by keywords, time range, and content type (audio, OCR, accessibility, input)
automatic Bearer token injection via SCREENPIPE_LOCAL_API_KEY environment variable
respects token limits by writing large API responses to files and sampling data intelligently
full access to Screenpipe's REST API for recordings, transcriptions, metadata, and analytics
relative (e.g., '1h ago') and ISO 8601 timestamp support for precise data slicing
target specific capture types (accessibility tree, audio, OCR, keystroke logs, memory)
iterate through large result sets with limit and offset parameters
all queries run against localhost:3030; no cloud sync or external API calls
Example Output
Example 1: Search screen activity from the last hour
User: "What apps did I use in the last hour?"
Claude queries: GET /search?q=&content_type=accessibility&limit=20&start_time=1h%20ago
Response: Array of 12 accessibility events showing Chrome (Gmail), VS Code (Python file edits), Slack conversations, and Calendar views with timestamps and UI element details.
Example 2: Find meeting transcript segments
User: "What was discussed in my 2pm meeting?"
Claude queries: GET /search?q=meeting&content_type=audio&start_time=2024-01-15T14:00:00Z&end_time=2024-01-15T15:00:00Z
Response: Audio transcription chunks with speaker labels, timestamps, and confidence scores for the Zoom meeting window.
Example 3: Analyze productivity patterns
User: "How much time did I spend coding vs. in meetings today?"
Claude queries multiple endpoints to aggregate accessibility events by app category, calculates duration per app, and returns a summary: "4.2 hours coding (VS Code, terminal), 1.8 hours meetings (Zoom, Calendar)."
What's Included
- SKILL.md: Complete Screenpipe API reference with authentication, endpoint documentation, and context window protection guidelines
- Search endpoint documentation: Parameters (q, content_type, limit, offset, start_time), examples, and modality explanations
- Authentication template: Ready-to-use Bearer token header format and environment variable setup
- Large response handling pattern: File I/O and jq sampling strategies to keep API queries within context limits
- Endpoint reference link: Full 60+ endpoint catalog at https://docs.screenpi.pe/llms-full.txt
Who It's For
- Software engineers — analyze debugging sessions, code review logs, and development tool usage patterns
- Product managers — track user session data and feature interaction telemetry from personal recordings
- Data analysts — export and process screen activity logs for productivity research and time tracking
- Researchers — study user behavior, UI interaction patterns, and application usage without external cloud infrastructure
- Productivity consultants — provide data-driven insights on client app usage, meeting time allocation, and focus periods
Best For
- Screen activity audits — retrieve what apps, files, and websites were accessed during specific time windows
- Meeting transcription & retrieval — search and extract audio transcriptions from recorded calls and conferences
- Productivity metrics — calculate time spent per application, meeting load, and deep work vs. interruption patterns
- Data export & compliance — pull structured screen logs for personal archives, audit trails, or compliance reviews
- AI-powered insights — feed screen context and transcriptions into Claude for intelligent summaries, action items, and analysis







