screenpipe-mcp 0.5.0 → 0.7.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -64,32 +64,42 @@ npx @modelcontextprotocol/inspector npx screenpipe-mcp
64
64
 
65
65
  ## Available Tools
66
66
 
67
- ### Cross-Platform
68
-
69
- - **search-content** - Search through recorded screen content, audio transcriptions, and UI elements
70
- - Full text search with content type filtering (OCR/Audio/UI)
71
- - Time range and app/window filtering
72
- - Pagination support
73
-
74
- - **pixel-control** - Control mouse and keyboard
75
- - Type text, press keys, move mouse, click
76
-
77
- ### macOS Only
78
-
79
- - **find-elements** - Find UI elements in applications by role
80
- - **click-element** - Click UI elements by accessibility ID
81
- - **fill-element** - Type text into UI elements
82
- - **scroll-element** - Scroll UI elements
83
- - **open-application** - Open applications by name
84
- - **open-url** - Open URLs in default browser
67
+ ### search-content
68
+ Search through recorded screen content (OCR) and audio transcriptions:
69
+ - Full text search with content type filtering (OCR/Audio/UI)
70
+ - Time range and app/window filtering
71
+ - Speaker filtering (by ID or name)
72
+ - Pagination support
73
+
74
+ ### search-ui-events (macOS)
75
+ Search UI input events captured via accessibility APIs. This is the third data modality alongside vision and audio:
76
+ - **Event types**: `click`, `text`, `scroll`, `key`, `app_switch`, `window_focus`, `clipboard`
77
+ - Filter by app, window, time range
78
+ - `text` events show aggregated keyboard input (what was typed)
79
+ - `click` events include accessibility element labels
80
+ - `clipboard` events show copy/paste content
81
+
82
+ ### get-ui-event-stats (macOS)
83
+ Get aggregated statistics of UI events:
84
+ - Event counts grouped by app and event type
85
+ - Useful for productivity analysis and app usage tracking
86
+
87
+ ### export-video
88
+ Export screen recordings as video files:
89
+ - Specify time range with start/end times
90
+ - Configurable FPS for output video
85
91
 
86
92
  ## Example Queries in Claude
87
93
 
88
94
  - "Search for any mentions of 'rust' in my screen recordings"
89
95
  - "Find audio transcriptions from the last hour"
90
96
  - "Show me what was on my screen in VSCode yesterday"
91
- - "Open Safari and go to github.com"
92
- - "Find the search button in Chrome and click it"
97
+ - "Export a video of my screen from 2-3pm today"
98
+ - "Find what John said in our meeting about the database"
99
+ - "What did I type in Slack today?" (uses search-ui-events)
100
+ - "Show me my app usage statistics for the past 3 hours"
101
+ - "What did I copy to clipboard recently?"
102
+ - "Which apps did I switch between most today?"
93
103
 
94
104
  ## Requirements
95
105