wispr
Control the Wispr Flow voice dictation app on macOS from the shell via its `wispr-flow://` URL scheme — lets an AI agent (e.g. Claude Code) turn Wispr Flow listening on and off for the user. Use when the user wants an agent to start or stop dictation programmatically, toggle hands-free listening, or switch the dictation microphone — e.g. "start dictation", "start wispr", "start listening", "turn on wispr", "stop listening", "stop dictation", "switch mic".
소스 정보
- 저장소
- Vesely/skills
- 최근 소스 활동
- 2026년 6월 26일 10:34
- 감지된 SKILL.md 언어
- 영어
- 스타
- 28
- 포크
- 4
설치 방법
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
소스 파일 검토
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
SKILL.md 표시 중
SKILL.md
소스 지침 · 읽기 전용 미리보기- name
- wispr
- description
- Control the Wispr Flow voice dictation app on macOS from the shell via its `wispr-flow://` URL scheme — lets an AI agent (e.g. Claude Code) turn Wispr Flow listening on and off for the user. Use when the user wants an agent to start or stop dictation programmatically, toggle hands-free listening, or switch the dictation microphone — e.g. "start dictation", "start wispr", "start listening", "turn on wispr", "stop listening", "stop dictation", "switch mic".
# wispr
Control [Wispr Flow](https://wisprflow.ai/) (the AI voice dictation app) from the command line on macOS via its registered `wispr-flow://` URL scheme. No keystroke simulation, no Accessibility permission, no API key — just `open`.
The main use case: let an AI coding agent (Claude Code, or any tool that can run a shell command) **turn Wispr Flow listening on and off for the user** — e.g. the user says "start dictation" and the agent fires the start deeplink so they can speak their next prompt hands-free.
Requires the Wispr Flow desktop app installed and running, with Microphone permission granted.
## Start listening (hands-free / toggle mode)
```bash
open "wispr-flow://start-hands-free"
```
Starts hands-free dictation — Flow keeps listening until stopped (unlike push-to-talk, which you hold). Dictated text is typed into **whatever window currently has keyboard focus**, so put the cursor where you want the text before triggering.
## Stop listening
```bash
open "wispr-flow://stop-hands-free"
```
## Switch microphone
```bash
open "wispr-flow://switch-mic?mic_name=<PREFIX>"
```
Switches to the first input device whose name starts with `<PREFIX>` (case-insensitive). Example:
```bash
open "wispr-flow://switch-mic?mic_name=MacBook"
```
## Behavior / gotchas
- **`start-hands-free` only starts when Flow is idle.** If dictation is already active it is silently ignored. To guarantee a clean start, send `stop-hands-free` first.
- **`stop-hands-free` only acts while in hands-free (locked) listening mode.** It will not cancel a push-to-talk session.
- **Text lands in the focused field.** Triggered from a terminal, Flow types into the terminal — switch focus to the target app first.
- **Undocumented / unofficial.** These deeplink actions are not in Wispr Flow's public docs; they were verified against app version **1.5.980** (Apple Silicon, macOS) and may change in future releases. The documented, supported activation path is the in-app hotkey (default **Fn** for push-to-talk, **Fn+Space** for hands-free), configurable under Settings → General → Shortcuts.
GitHub에서 보기