dsh-speech-coach
DSH plugin bundle that puts the speech-coach MCP server in your chat
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/chxjames--dsh-speech-coach)HTML
<a href="/listings/chxjames--dsh-speech-coach"><img src="/api/listings/chxjames--dsh-speech-coach/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.
Install the bundle directory from Plugin Manager (the same operation is `plugin_manager install_bundle` with this folder as the target). Windows is wired up already (`cmd /c npx …`); on macOS or Linux edit the two lines in `cordis.patch.yml` as the comment there says.
After installing, restart Harness once so the new MCP row is loaded — then the tool shows up as `mcp__speech-coach__analyze_speech`.
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.
README.md · README · 40 lines
- # dsh-speech-coach
- A DSH plugin that puts [speech-coach](https://www.npmjs.com/package/speech-coach) in your chat.
- Say *"score ~/talks/rehearsal.mp4 as a competition talk"* and you get the same report the web page
- downloads: a score out of 100, the findings ordered by points lost, a sentence-by-sentence table and the
- transcript. Nothing is uploaded — the audio is decoded and transcribed on this machine.
- ## What it adds
- | | |
- |---|---|
- | Tool | `mcp__speech-coach__analyze_speech` |
- | Server | `npx -y speech-coach` (stdio) |
- | Needs | Node 20+, a video file on disk |
- | Network | the package and ~80 MB of Whisper weights on the first run, then nothing |
- ## Install
- Install the bundle directory from Plugin Manager (the same operation is
- `plugin_manager install_bundle` with this folder as the target). Windows is
- wired up already (`cmd /c npx …`); on macOS or Linux edit the two lines in
- `cordis.patch.yml` as the comment there says.
- After installing, restart Harness once so the new MCP row is loaded — then the
- tool shows up as `mcp__speech-coach__analyze_speech`.
- ## Use
- > Analyse `E:\talks\rehearsal-final.mp4` as a 3MT contest talk at solid level and tell me the one thing to fix.
- Useful arguments: `scene` (`homework`, `final`, `contest`, `interview`, `defense`, `pitch`), `level`
- (`starter`, `solid`, `pro`), `format` (`camera`, `stage`), `lang` (`en`, `zh`), `content` and `composition`
- (your own 0–100 scores for the two things a machine cannot judge).
- ## Notes
- - The first tool call is slow: it installs the package and downloads the model. Later calls reuse both.
- - Long talks: a 3-minute video takes about a minute on CPU. The bundle allows 30 minutes per call.
- - This bundle only declares the connection; the engine and the scoring live in the published package.
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.