eddiesanjuan/markupr

View on GitHub ↗

Give your AI coding agent eyes and ears. Screen + voice capture → structured Markdown. MCP server, CLI, and macOS app.

18 ★3 forksTypeScriptUpdated 4mo ago

What you need to know

Desktop capture app and MCP server that turns screen recordings with voice narration into structured Markdown reports with extracted frames for AI coding agents.

Install

Desktop app from markupr.com or Releases
Or MCP: npx --package markupr markupr-mcp
Requires ffmpeg on PATH

Usage

  • •Record + narrate + paste report path into agent
  • •markupr analyze ./recording.mov
  • •markupr watch ~/Desktop

Key features

  • ✓Screen recording with hotkey and narration
  • ✓Whisper transcription and frame extraction at exact timestamps
  • ✓Cursor/window/focus context attached to shots
  • ✓CLI for existing recordings
  • ✓GitHub Action for PR feedback

Caveats

  • ⚠Local-first: Whisper runs on device
  • ⚠macOS notarization rolling out (may need xattr workaround)
Platforms: macOS · WindowsClients: Claude Code · Cursor · Windsurf

Reviewed 2026-08-11

Topics

ai-agentsai-toolsclicode-reviewcursordeveloper-toolselectrongithub-actionllmmcpmcp-servermodel-context-protocolscreen-recordingscreenshottranscriptiontypescriptvisual-feedbackvoice-recordingwhisperwindsurf
Stars
18★
Forks
3
Language
TypeScript
License
MIT
Created
2026-02-04
Last push
2026-05-01