Skip to content
Back to search

vedivox

MarketplaceWorks with Claude CodeGitHub

Record and transcribe meetings on your Mac, locally

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/dimstunt--vedivox/badge.svg)](/listings/dimstunt--vedivox)

HTML

<a href="/listings/dimstunt--vedivox"><img src="/api/listings/dimstunt--vedivox/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

brew install --cask dimstunt/tap/vedivox

The cask installs `Vedivox.app` and links the `vedivox` CLI and the `vedivox-mcp` server into Homebrew's `bin`. The Claude Code skills are not part of the cask: they install as a plugin, two commands under [Command line and MCP](#command-line-and-mcp). Your meetings are never removed by `brew uninstall --zap`.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 14 lines
  1. {
  2. "name": "vedivox",
  3. "owner": { "name": "Dmitrii Cherkashin" },
  4. "metadata": {
  5. "description": "Claude Code skills for Vedivox, the macOS meeting recorder"
  6. },
  7. "plugins": [
  8. {
  9. "name": "vedivox",
  10. "source": "./",
  11. "description": "Ask about your Vedivox meetings, get help during a call from its live text, and give a meeting a title and a timestamped description"
  12. }
  13. ]
  14. }
README.md · README · 243 lines
  1. # <picture><source media="(prefers-color-scheme: dark)" srcset="docs/images/vedivox-header-dark.svg"><img alt="Vedivox" src="docs/images/vedivox-header-light.svg" height="64"></picture>
  2. [![test](https://github.com/dimstunt/vedivox/actions/workflows/test.yml/badge.svg)](https://github.com/dimstunt/vedivox/actions/workflows/test.yml)
  3. [![GitHub stars](https://img.shields.io/github/stars/dimstunt/vedivox?style=flat)](https://github.com/dimstunt/vedivox/stargazers)
  4. [![Licence: GPL-3.0-or-later](https://img.shields.io/badge/licence-GPL--3.0--or--later-blue)](LICENSE)
  5. A macOS meeting recorder that transcribes and tells speakers apart on your Mac.
  6. If Vedivox helps you, a ⭐ helps others find it.
  7. ![A meeting in Vedivox: the transcript by speaker, the player and the meeting notes](docs/images/transcript.png)
  8. ![Live text of a call being recorded: lines arrive as they are said, then the voices Vedivox has met get their names](docs/images/screencast-live.gif)
  9. Vedivox records your microphone and the system audio of a call (Zoom, Google Meet,
  10. Microsoft Teams, Yandex Telemost, anything that plays sound), turns both into a transcript, and labels who said
  11. what. Voices it has met before are recognized by name in later meetings. It
  12. also types dictation into any app (press a hotkey, speak, press it again), imports existing recordings, and
  13. exposes everything through a command line and an MCP server, so an assistant
  14. such as Claude Code can search and summarize your meetings.
  15. Status: pre-1.0, built for one person's daily use and shared as is. Database
  16. migrations may still break backward compatibility between versions.
  17. ## Requirements
  18. - macOS 14.4 or later on Apple Silicon (the models run on the Neural Engine).
  19. - Internet on first launch: the speech models are downloaded once (about 1.2 GB).
  20. After that, recording, transcription and speaker recognition work offline.
  21. - About 3 GB of free disk for the models and their compiled caches.
  22. Transcription uses NVIDIA Parakeet TDT 0.6B v3, which covers 25 European languages:
  23. Bulgarian, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German,
  24. Greek, Hungarian, Italian, Latvian, Lithuanian, Maltese, Polish, Portuguese, Romanian,
  25. Russian, Slovak, Slovenian, Spanish, Swedish and Ukrainian (per the model card); other
  26. languages are not supported.
  27. <p>
  28. <img src="docs/images/sidebar.png" alt="The sidebar: meetings by day, with series and summaries" width="260">
  29. <img src="docs/images/speakers.png" alt="Speakers: voices Vedivox has learned, recognized by name in later meetings" width="520">
  30. </p>
  31. ## Install with Homebrew
  32. ```bash
  33. brew install --cask dimstunt/tap/vedivox
  34. ```
  35. The cask installs `Vedivox.app` and links the `vedivox` CLI and the `vedivox-mcp` server into Homebrew's
  36. `bin`. The Claude Code skills are not part of the cask: they install as a plugin, two commands
  37. under [Command line and MCP](#command-line-and-mcp). Your meetings are never removed by
  38. `brew uninstall --zap`.
  39. ## Install from a release zip
  40. Unpack the archive and run `bash install.sh` in it. The full walkthrough —
  41. permissions, the CLI and the Claude Code skills — is in
  42. [docs/INSTALL.md](docs/INSTALL.md).
  43. ## Build from source
  44. You need Xcode 26 or later and [XcodeGen](https://github.com/yonaskolb/XcodeGen)
  45. (`brew install xcodegen`).
  46. ```bash
  47. make install TEAM=- # ad-hoc signature, no Apple account needed
  48. make install TEAM=ABCDE12345 # or sign with your own Apple Developer team
  49. ```
  50. `make install` builds a Release `Vedivox.app` into `/Applications` and puts `vedivox`
  51. and `vedivox-mcp` into `~/.local/bin`. Without `TEAM` the build signs with the
  52. author's team from `Vedivox/project.yml` when its certificate is in the keychain and
  53. ad hoc otherwise (the Makefile says so in its first line of output). Dependency
  54. versions are pinned by the committed `Package.resolved` files.
  55. With an ad-hoc signature (`TEAM=-`) macOS treats every rebuild as a new app and
  56. asks for the microphone and system-audio permissions again. A real team ID keeps
  57. them across rebuilds.
  58. The same `TEAM=` works for the tests:
  59. ```bash
  60. make test TEAM=- # headless logic suite, the gate
  61. make test-ui TEAM=- # HUD snapshot tests, needs a full app build
  62. ```
  63. `Vedivox/Vedivox.xcodeproj` is generated from `Vedivox/project.yml` (`make generate`); do
  64. not edit it by hand. The benchmark target (`make bench`) and a few recorded
  65. scenarios rely on private recordings that are not in this repository; those
  66. tests skip themselves.
  67. `make dist` packs `Vedivox.app`, the CLI and the docs into a zip for another Mac.
  68. `make dist-notarized DEVELOPER_ID="Developer ID Application: …"` signs and
  69. notarizes that zip (see the comment in the `Makefile`); without `DEVELOPER_ID` it takes
  70. the Developer ID certificate from the keychain.
  71. ## Command line and MCP
  72. `vedivox` reads the same database as the app and sends actions to the running app
  73. over a local Unix socket: `vedivox list`, `vedivox show <id>`, `vedivox transcript <id>`,
  74. `vedivox search <query>`, `vedivox speakers`, `vedivox import <file|url>`,
  75. `vedivox transcribe <file|url>`, `vedivox reprocess <id> --yes`, `vedivox record start|stop`,
  76. `vedivox live` (the text of the meeting being recorded right now),
  77. `vedivox meeting split|trim|merge|undo` (cut a long recording into calls, join a call recorded in parts),
  78. and editing commands for speakers, series, marks and segments. The reading commands,
  79. `transcribe`, `live`, `record`, `meeting split|trim|merge|undo|delete`, `audio compact` and
  80. `feedback` take `--json`; the rest print text only. `vedivox --help` lists them all.
  81. Ready-made Claude Code skills live in
  82. [skills/vedivox](skills/vedivox), [skills/vedivox-live](skills/vedivox-live) (help during a
  83. call) and [skills/vedivox-enrich](skills/vedivox-enrich) (a title and a timestamped description
  84. for a meeting). This repository is also a Claude Code plugin marketplace, so the skills install
  85. and update as a plugin. In Claude Code:
  86. ```
  87. /plugin marketplace add dimstunt/vedivox
  88. /plugin install vedivox@vedivox
  89. ```
  90. They are then `/vedivox:vedivox`, `/vedivox:vedivox-live` and `/vedivox:vedivox-enrich`, and
  91. `/plugin marketplace update vedivox` brings the newer ones after a release. `install.sh` copies
  92. the same skills into `~/.claude/skills` instead; use one way or the other.
  93. `vedivox-mcp` is a Model Context Protocol server over stdio. It lists, reads and searches
  94. meetings (the live transcript of a recording too), imports files and YouTube links,
  95. reprocesses, splits, trims and merges meetings and undoes those edits, renames and merges
  96. speakers, reassigns lines, sets a meeting's title, description and series, creates series, and
  97. adds and removes excluded intervals. It does not record, and has no counterpart of
  98. `transcribe`, `status`, `queue`, `audio compact`, `feedback`, `mute-debug`, `meeting delete`,
  99. `speaker detach|delete` or `series set|merge|delete|apply-rules`. Its 22 tools:
  100. `list_meetings`, `get_meeting`, `get_transcript`, `get_live_transcript`, `search_transcripts`,
  101. `list_speakers`, `update_speaker`, `merge_speakers`, `update_meeting`, `update_segment`,
  102. `reprocess_meeting`, `split_meeting`, `trim_meeting`, `merge_meetings`, `undo_meeting_edit`,
  103. `import_file`, `import_youtube`, `list_series`, `create_series`, `mark_excluded`, `unmark`,
  104. `list_marks`.
  105. Point an MCP client at the binary by its full path:
  106. ```json
  107. {
  108. "mcpServers": {
  109. "vedivox": { "command": "/Users/you/.local/bin/vedivox-mcp" }
  110. }
  111. }
  112. ```
  113. That is `~/.local/bin/vedivox-mcp` after `install.sh` or `make install`, and
  114. `/opt/homebrew/bin/vedivox-mcp` (Homebrew's `bin`) after the cask. With Claude Code the
  115. CLI plus the skill is usually enough, and MCP is optional.
  116. Who can drive the app: any process running as your macOS user, not just `vedivox`
  117. and `vedivox-mcp`. The socket does not check which program connects; other users on
  118. the Mac cannot reach it. Such a process can start a recording and read the
  119. audio, so Vedivox shows a notification, "Recording started from the command line",
  120. whenever a recording is started through the socket. See
  121. [SECURITY.md](SECURITY.md).
  122. ## Privacy
  123. Recording, transcription and speaker recognition run on the Mac with Core ML.
  124. Audio, transcripts, voice profiles and the database stay in
  125. `~/Library/Application Support/Vedivox/`. There is no account, telemetry, analytics,
  126. crash reporting or auto-install, and the code calls no cloud or LLM API. The MCP
  127. server speaks stdio and the app's control channel is a Unix socket readable
  128. only by you; nothing listens on the network.
  129. The exceptions, so the claim stays honest:
  130. - On first launch (and again if the model cache is cleared) the speech models
  131. are downloaded from Hugging Face over HTTPS. After that Vedivox works offline.
  132. - `vedivox import <url>` and `vedivox transcribe <url>` run the `yt-dlp` you installed
  133. yourself, which contacts the video site.
  134. - Once a day Vedivox asks the GitHub API for its latest release, to tell you when a
  135. newer version is out. It is one anonymous GET; nothing about you or your
  136. meetings is sent, and nothing is downloaded or installed. Turn it off in
  137. Settings → About.
  138. - `vedivox feedback` writes a zip with logs to your Desktop. Nothing is sent, but
  139. the logs contain meeting titles, speaker names and file paths — read it before
  140. you share it.
  141. - Calendar access uses the local macOS calendar store. Reading the browser's
  142. tabs (to notice a Google Meet or Yandex Telemost call) happens locally through Apple Events.
  143. - The app is not sandboxed, so all of the above is a property of the code, not
  144. something macOS enforces.
  145. ## Recording other people
  146. In many countries and US states, recording a conversation requires the consent
  147. of everyone in it, sometimes announced in advance. Vedivox does not tell anyone
  148. that it is recording: it shows its own recording indicator on your Mac and
  149. nothing else, and it records whatever your Mac plays and hears. Knowing the
  150. rules where you and the other people are, and asking for consent, is your
  151. responsibility. This is not legal advice.
  152. ## Documentation
  153. - [docs/GUIDE.md](docs/GUIDE.md) — the non-obvious parts of the UI: hotkeys,
  154. modifier clicks, settings files. Also in the app under Help → Vedivox Guide.
  155. - [CHANGELOG.md](CHANGELOG.md) — what changed in each version.
  156. - [docs/INSTALL.md](docs/INSTALL.md) — installing a release zip.
  157. ## Acknowledgements
  158. Vedivox is glue around other people's work.
  159. Libraries:
  160. | Library | Licence | Used in |
  161. |---|---|---|
  162. | [FluidAudio](https://github.com/FluidInference/FluidAudio) (includes fastcluster, BSD-2-Clause, and a port of VBx, Apache-2.0) | Apache-2.0 | app |
  163. | [GRDB.swift](https://github.com/groue/GRDB.swift) | MIT | app, CLI, MCP server |
  164. | [KeyboardShortcuts](https://github.com/sindresorhus/KeyboardShortcuts) | MIT | app |
  165. | [swift-argument-parser](https://github.com/apple/swift-argument-parser) | Apache-2.0 with Runtime Library Exception | CLI |
  166. | [swift-snapshot-testing](https://github.com/pointfreeco/swift-snapshot-testing) and its dependencies | MIT; swift-syntax is Apache-2.0 | tests only |
  167. Models. None are stored in this repository or in the app bundle; FluidAudio
  168. downloads them at run time from the Fluid Inference repositories on Hugging
  169. Face. All of them are Core ML conversions, that is, modified versions of the
  170. originals.
  171. | Used for | Downloaded from | Original model | Licence |
  172. |---|---|---|---|
  173. | Transcription | [FluidInference/parakeet-tdt-0.6b-v3-coreml](https://huggingface.co/FluidInference/parakeet-tdt-0.6b-v3-coreml) | NVIDIA [parakeet-tdt-0.6b-v3](https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3) | [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/) |
  174. | Speaker diarization and voice embeddings | [FluidInference/speaker-diarization-coreml](https://huggingface.co/FluidInference/speaker-diarization-coreml) | [pyannote speaker-diarization-community-1](https://huggingface.co/pyannote/speaker-diarization-community-1), with WeSpeaker embeddings and PLDA from BUT Speech@FIT; the legacy path uses pyannote segmentation-3.0 (MIT, © CNRS) and wespeaker-voxceleb-resnet34-LM | CC-BY-4.0 |
  175. | Live text while dictating | [FluidInference/Nemotron-3.5-ASR-Streaming-Multilingual-0.6b-CoreML](https://huggingface.co/FluidInference/Nemotron-3.5-ASR-Streaming-Multilingual-0.6b-CoreML) | NVIDIA nemotron-3.5-asr-streaming-0.6b | [OpenMDW-1.1](https://openmdw.ai/license/1-1/) |
  176. Thanks to NVIDIA, the pyannote and WeSpeaker authors, BUT Speech@FIT and Fluid
  177. Inference. The licences above are those stated on the model cards in
  178. November 2026; the cards are the authority if they change.
  179. The licences and notices of everything built into the app, the CLI and the MCP server, in
  180. full, are in [THIRD_PARTY_NOTICES.md](THIRD_PARTY_NOTICES.md), which also ships in the
  181. release zip.
  182. URL import relies on [yt-dlp](https://github.com/yt-dlp/yt-dlp) and
  183. [ffmpeg](https://ffmpeg.org), which Vedivox does not ship: install them yourself
  184. (`brew install yt-dlp ffmpeg`) if you want that feature.
  185. ## Licence
  186. Vedivox is free software under the [GNU General Public License, version 3 or any
  187. later version](LICENSE) (GPL-3.0-or-later). © 2026 Dmitrii Cherkashin
  188. Issues, ideas and pull requests are welcome; a pull request is merged under the
  189. contributor licence agreement in [CONTRIBUTING.md](CONTRIBUTING.md).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.