call.md  by video-db

Meeting intelligence and agent loops for live calls

Created 2 months ago
332 stars

Top 82.5% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

Call.md transforms meetings into actionable intelligence loops, enabling users to record, transcribe, and analyze conversations with real-time AI. Designed for professionals seeking to enhance meeting productivity and integrate insights into workflows, it offers live assistance during calls and automated post-meeting summaries and data exports.

How It Works

The application captures dual-channel audio (speaker vs. system) locally, streaming it to VideoDB for real-time transcription via WebSocket. It simultaneously processes live intelligence, including conversation metrics, coaching nudges, and AI-driven suggestions. A Model Context Protocol (MCP) agent automatically detects information needs and triggers integrated tools, with results displayed inline. Post-meeting, it generates multi-faceted AI summaries and structured exports, facilitating workflow automation via webhooks.

Quick Start & Requirements

  • Install: curl -fsSL https://artifacts.videodb.io/call.md/install | bash
  • Prerequisites: macOS 12+ (Monterey or later), VideoDB API Key (free tier available), Microphone and Screen Recording system permissions. Linux support is planned.
  • Docs: docs.videodb.io

Highlighted Details

  • Live Assist: Provides contextual AI suggestions and talking points during live calls.
  • MCP Auto-Triggering: Dynamically invokes external tools based on conversational context and information needs.
  • AI-Generated Summaries: Delivers narrative overviews, topic-based key points, and actionable items post-meeting.
  • Local-First Storage: Utilizes SQLite for on-device data storage, enhancing privacy.

Maintenance & Community

Developed by the VideoDB team. Community support is available via Discord. Comprehensive documentation is hosted at docs.videodb.io.

Licensing & Compatibility

The README does not specify a software license. Compatibility is currently limited to macOS and Windows; Linux support is forthcoming.

Limitations & Caveats

Linux operating systems are not yet supported. Real-time transcription and AI intelligence features necessitate an active internet connection and a valid VideoDB API key. The absence of a stated software license may pose adoption risks.

Health Check
Last Commit

2 weeks ago

Responsiveness

Inactive

Pull Requests (30d)
1
Issues (30d)
0
Star History
34 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.