Menu
Transcribe audio files using Mistral AI Voxtral.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
ABCd context protocol across ABC root files (AGENTS.md, BACKLOG.md, CHANGELOG.md), human entrypoint README.md files tree, and /docs. Use after meaningful project changes, backlog drift, documentation refactors, self-evolving context maintenance, or a forced context reconciliation pass.
Build, refactor, review, or debug Svelte 5 components that use Bits UI primitives. Use when working with bits-ui dialogs, popovers, dropdowns, comboboxes, selects, tabs, date/time controls, menus, tooltips, portals, render delegation, or Bits UI type helpers.
Collaborative idea-to-design and inquiry protocol. Use for product/architecture exploration, research-style question shaping, feature design, standards, specs, UX concepts, module boundaries, and non-trivial behavior changes when uncertainty matters.
Pragmatic senior-engineer operating contract for coding agents. Use for coding, implementation, refactoring, debugging, code review, tests, validation, repository maintenance, file edits, or technical investigation.
Horizontal Gene Transfer protocol for skills. Synchronizes best practices and architectural patterns across the skill library.
Manage a guarded release flow that commits prepared release work on dev, opens a dev-to-main pull request with a release-focused PR summary, waits for checks, merges on success, tags, and optionally publishes an existing npm package. Use when the user asks to prepare or execute a dev→main release PR, hotfix release PR, or Dev2Main PR Summary workflow.
| name | mistral-stt |
| description | Transcribe audio files using Mistral AI Voxtral. |
| metadata | {"version":"1.0.18"} |
Standalone direct Node.js client for Mistral's Voxtral transcription API. The canonical client is scripts/transcribe.mjs; scripts/transcribe.sh is the shell entrypoint wrapper. There are no curl fallbacks or Python parser dependencies.
MISTRAL_API_KEY=xxx ./scripts/transcribe.sh audio.ogg [language] [model]
MISTRAL_API_KEY=xxx ./scripts/transcribe.sh --file audio.ogg --lang ru --model voxtral-mini-latest
language — optional; omitted means provider auto-detection.model — optional; default: voxtral-mini-latest.MISTRAL_API_KEY is missing.--file, -f — audio file path.--lang, --language, -l — optional language code.--model, -m — Mistral transcription model.--help, -h — usage.fetch, FormData, and Blob.voxtral-mini-latest.text field.transcribe_mistral: transcribe.sh {file} {lang} {model}.