silverstein/minutes

latest · 0.26.3

The private, owned conversation-memory layer for AI. Record, transcribe, and search every meeting.

GitHub: silverstein/minutes
npm: minutes-mcp
MCP registry: io.github.silverstein/minutes

Install

claude mcp add io-github-silverstein-minutes -- npx minutes-mcp

/mcp confirms it's connected.

Install

Or add it manually in ~/.cursor/mcp.json:

{
  "mcpServers": {
    "io-github-silverstein-minutes": {
      "command": "npx",
      "args": [
        "minutes-mcp"
      ]
    }
  }
}

In claude_desktop_config.json (Settings → Developer → Edit Config), then restart:

{
  "mcpServers": {
    "io-github-silverstein-minutes": {
      "command": "npx",
      "args": [
        "minutes-mcp"
      ]
    }
  }
}

Config entry, any MCP client:

{
  "command": "npx",
  "args": [
    "minutes-mcp"
  ]
}

Community score

Community66/ 100

from web signals · no reviews yet · 3 mentions · 3 reactions (community v3)

Mentions (reaction sentiment)100/100 · 12% of the weight
Web signals (ecosystem baseline)62/100 · 88% of the weight

Used it? Rate v0.26.3

No reviews yet — be the first.

What the web says

Ecosystem baseline62/ 100

Objective public usage & activity (formula v3) — the community score's web-signals input, and its own sort.

GitHubsource ↗

GitHub stars · 15 days ago: ★ 1.5K
Last commit · 15 days ago: 1 day ago
Issue close rate · 15 days ago: 76%
Release cadence · 15 days ago: about every 5 days
Latest release · 15 days ago: v0.27.2
License · 15 days ago: MIT

Packages

No package-registry data collected yet.

Listed on

Not yet found on other registries or list sites.

Mentions

Quotes3

Opt-in auto-record / auto-stop on detected calls (Microsoft Teams first) ↗
Ask Opt-in automatic recording driven by call detection: when a detected call starts (Microsoft Teams desktop first, since it is already in the default call_detection.apps alongside Zoom and Webex), start a call-intent recording without a c…
GitHub issue · @silverstein · 1 reaction · 100% positive
Speaker attribution can override identity.name with a name discussed in transcript ↗
Summary When the user has `identity.name` correctly set in `config.toml`, the diarization-to-speaker-name mapping can still misidentify them by picking up a different name mentioned in the transcript content and assigning it to their speake…
GitHub issue · @silverstein · 1 reaction · 100% positive
Support for Vulkan and ROCm/HIP GPU backends ↗
whisper-rs 0.16 exposes vulkan and hipblas feature flags (from whisper.cpp), but minutes doesn't wire them through yet. Adding these would let AMD GPU users (and anyone with Vulkan support) get hardware-accelerated transcription. The change…
GitHub discussion · @mwolson · 1 reaction · 100% positive

Docs & liveness

No liveness or verification signals yet.

Version history1

0.26.3release date unknown

Badge for your README

Shows this page's community score and its sample, kept current, and links readers here to review.

[![rateme12 community score](https://api.rateme12.com/servers/578395d2-cc57-4071-aba5-626acbddcfd6/badge.svg)](https://rateme12.com/servers/578395d2-cc57-4071-aba5-626acbddcfd6)

Details

Transport: stdio
Status: listed
Source: Imported from the MCP registry
Repository: source ↗