Voice Buddy

CC — Your personality-driven voice companion for Claude Code.

Voice Buddy turns Claude Code into a more human experience. CC is a personality-driven voice companion that speaks at key moments in your coding workflow.

Open Claude Code       → "欢迎回来，今天也要加油哦~"         (cute-girl)
Needs your attention   → "Boss, your attention is needed"    (secretary)
Task completed         → "Senpai、バグ直したよ~"              (kawaii)
Close session          → "辛苦了，好好休息一下"               (warm-boy)

7 Styles, 3 Languages

Style	Language	Description	TTS Voice	Default Nickname
cute-girl	Chinese	Cute, sweet	zh-CN-XiaoyiNeural	Master
elegant-lady	Chinese	Graceful, intellectual	zh-CN-XiaoxiaoNeural	Master
warm-boy	Chinese	Warm, caring	zh-CN-YunxiNeural	Master
secretary	English	Professional, efficient	en-US-JennyNeural	Boss
steward	English	British butler, composed	en-GB-RyanNeural	Sir
cyber-girl	English	Cyberpunk femme, cold	en-GB-SoniaNeural	Commander
kawaii	Japanese	Cute, energetic	ja-JP-NanamiNeural	Senpai

How It Works

Voice Buddy hooks into Claude Code's hook system, playing audio at key events.

Three-tier audio architecture:

Tier	Latency	Used For	How
Pre-packaged	<100ms	Session start/end	Pre-generated MP3 playback
Real-time TTS	~1s	Notifications	edge-tts synthesis with nickname substitution
AI Summary	~3s	Task completion	Claude subagent generates summary + TTS

Installation

Prerequisites

Claude Code — tested against v2.1.179
Python 3.10 – 3.14 (3.9 is not supported; see Development below)
Audio player (macOS: afplay built-in, Linux: paplay/aplay/mpg123)

Claude Code compatibility. Voice Buddy ships its skill under skills/voice-buddy/SKILL.md and launches its hooks in exec form (command + args) resolved through ${CLAUDE_PLUGIN_ROOT}. Both are the current plugin contract — the flat commands/ layout and shell-form hooks that earlier releases used still load, but are no longer the documented default. Exec form and ${CLAUDE_PLUGIN_ROOT} substitution are verified on v2.1.179; the docs publish no explicit version floor for either, so older releases are untested rather than known-broken. Release validation is claude plugin validate . --strict.

Install from Plugin Marketplace

Run these two commands in Claude Code, one after the other:

/plugin marketplace add luyao618/voice-buddy   # 1. Add plugin source
/plugin install voice-buddy                                 # 2. Install plugin

During installation you'll be prompted to choose a style and nickname. Press Enter to use defaults (cute-girl / Master).

Hooks are auto-registered on install — no manual setup needed.

Upgrade

/plugin marketplace update voice-buddy-marketplace   # refresh the source
/plugin update voice-buddy@voice-buddy-marketplace   # apply the new version

Restart Claude Code afterwards. A listener left over from the previous version is terminated and replaced, so exactly one stays running.

Use the fully-qualified voice-buddy@voice-buddy-marketplace. The bare name reports Plugin "voice-buddy" not found.

Uninstall

/plugin uninstall voice-buddy

Your settings are kept on purpose, so reinstalling restores your style and nickname. To remove them as well:

# macOS
rm -rf ~/Library/Application\ Support/voice-buddy
# Linux
rm -rf ~/.config/voice-buddy

That directory holds config.json, the debug log, and the hotkey listener's runtime state — nothing else on your system is touched.

Recommended: Add Permission Allowlist

Add the following rules to the allow list in .claude/settings.json to skip permission prompts for voice playback:

"Bash(PYTHONPATH=* python3 -m voice_buddy*)",
"Bash(PYTHONPATH=* python3 -m voice_buddy.*)"

Configuration

After installation, type /voice-buddy in Claude Code to open the configuration panel.

You can also tell Claude what you want in natural language:

"Switch Voice Buddy to kawaii style"
"Change my nickname to Senpai"
"Disable notification voice"

Or use CLI commands (the voice-buddy command is automatically available after plugin install):

# Switch style
voice-buddy config --style kawaii

# Change nickname
voice-buddy config --nickname Senpai

# Set both at once
voice-buddy config --style secretary --nickname Boss

# Disable/enable specific events
voice-buddy config --disable notification
voice-buddy config --enable notification

# Global on/off
voice-buddy on
voice-buddy off

# Test
voice-buddy test sessionstart
voice-buddy test notification

Config file location:

macOS: ~/Library/Application Support/voice-buddy/config.json
Linux: ~/.config/voice-buddy/config.json
Windows: %APPDATA%\voice-buddy\config.json

Platform support. macOS and Linux are covered by CI. Windows has a config path and the hooks are written to work there, but it is untested — the global hotkey is macOS-only, and parts of the process supervision rely on POSIX primitives. Treat Windows as best-effort.

Supported Events

Event	When	Audio Source
SessionStart	Open Claude Code	Pre-packaged MP3
SessionEnd	Close session	Pre-packaged MP3
Notification	Claude sends a notification	Real-time TTS (with nickname)
Stop	Claude finishes a task	AI-generated summary

🔇 Global Hotkey to Stop Current Playback (macOS)

Press F2 at any time to immediately silence whatever Voice Buddy is currently playing — useful when a colleague walks over, you take a phone call, or a meeting starts unexpectedly.

One-time setup (required after first install)

The marketplace installer copies plugin files but does not install Python dependencies or grant macOS permissions. You need to do these three things once:

Step 1 — Install the macOS dependency

pip3 install 'pyobjc-framework-Quartz>=10.0,<13'

If you have multiple Python interpreters (system, Homebrew, pyenv, conda, depot_tools), make sure to install into the same one that runs voice-buddy. To find out which one:

voice-buddy hotkey-doctor --non-interactive
# Look for the row: [OK] python interpreter   current=/path/to/python3

Install pyobjc into that exact interpreter:

/path/to/python3 -m pip install 'pyobjc-framework-Quartz>=10.0,<13'

Step 2 — Enable "Use F1, F2, etc. as standard function keys"

Open System Settings → Keyboard → Keyboard Shortcuts… → Function Keys and turn ON "Use F1, F2, etc. keys as standard function keys".

Without this, F2 is intercepted by macOS as "brightness up" before any app can see it.

Step 3 — Grant Accessibility to your Python interpreter

Run the doctor to find the exact path:

voice-buddy hotkey-doctor --non-interactive
# Look for: [FAIL] Accessibility granted   grant Accessibility to: /path/to/python3

Then open System Settings → Privacy & Security → Accessibility:

Click the + button.
Press ⌘ + Shift + G to open "Go to Folder", paste the directory of that python path, hit Enter.
If python3 is a symlink (common with depot_tools / pyenv shims), it cannot be selected directly. Instead select the real binary it points to (e.g. python3.11). To check:
```
ls -la /path/to/python3
```
If macOS still refuses to select the file, open the directory in Finder (open /path/to/bin) and drag the python binary into the Accessibility list instead.
Make sure the toggle next to the new entry is green / on.

Step 4 — Restart Claude Code

Close and reopen Claude Code so SessionStart fires and spawns the listener subprocess.

Verify everything works

voice-buddy hotkey-doctor

Expected (interactive mode will ask you to press F2):

[OK  ] python interpreter
[OK  ] pyobjc importable
[OK  ] Accessibility granted  /usr/local/bin/python3 (EventTap created)
[OK  ] F-key fn-mode          F2 keydown observed
[OK  ] coord.lock writable
[OK  ] listener liveness      pid=...
[OK  ] version handshake
[OK  ] sessions registry
[OK  ] playback_pids sanity

If any row says FAIL or WARN, the detail message tells you what to fix.

Daily use

# Stop without using the hotkey (e.g. from another terminal)
voice-buddy stop

# Restart the hotkey listener (stops it by PID; a new session respawns it)
voice-buddy hotkey-restart

# Change the bound key
voice-buddy config --hotkey F3

# Disable / re-enable the hotkey listener
voice-buddy config --disable-hotkey
voice-buddy config --enable-hotkey

# Diagnose
voice-buddy hotkey-doctor
voice-buddy hotkey-doctor --non-interactive   # CI-friendly, skips the F2 press
voice-buddy hotkey-doctor --json              # machine-readable

How the listener lifecycle works

The listener is a singleton subprocess. One per machine, no matter how many Claude Code windows you open.
Spawn: SessionStart hook checks if a live listener exists; if not, spawns one (detached, ~500 ms cold start).
Self-exit: When the last Claude Code session ends, the listener notices via a 30 s idle timer and exits cleanly.
F2 = stop currently playing audio only. Queued TTS that hasn't started playing yet is not cancelled.
See docs/manual-tests.md for the full QA checklist.

Troubleshooting

Symptom	Cause	Fix
`voice-buddy hotkey-doctor` says `[FAIL] pyobjc importable`	pyobjc not installed in the right interpreter	Run Step 1 against the exact path the doctor reports
`[FAIL] Accessibility granted` after granting it	Granted the wrong interpreter, or the path is a symlink	Re-run doctor; grant the path it reports; if symlink, grant the real binary
F2 still doesn't work after all OK rows	Listener was spawned before pyobjc was installed	`voice-buddy hotkey-restart`, then start a new session
F2 stopped working and the doctor shows a listener that isn't running	The listener was killed uncleanly and its PID was recycled onto another process	`voice-buddy hotkey-restart` clears the stale record; a new session respawns it
F2 doesn't work after upgrading Python / recreating venv	Accessibility grant is bound to the executable path; the new python is a different file	Doctor will show `[WARN] python interpreter` with `DRIFT`; re-grant Accessibility to the new path
Want to verify in a CI pipeline	—	`voice-buddy hotkey-doctor --non-interactive --json`

Architecture

┌──────────────────────────────────────────────────────────┐
│                     Claude Code                          │
│                                                          │
│  Hook Event ──► voice_buddy (stdin JSON)                 │
│                      │                                   │
│            ┌─────────┼─────────┐                         │
│            │         │         │                         │
│     SessionStart  Notification  Stop                     │
│     SessionEnd       │         │                         │
│            │         │    injector.py                     │
│     Pre-packaged  Real-time   (exit code 2)              │
│       MP3 file    TTS via      │                         │
│            │     edge-tts   voice-buddy-*                │
│            │         │      subagent                     │
│         player.py  player.py  (haiku)                    │
│            │         │         │                         │
│         🔊 <100ms  🔊 ~1s   subagent_tts.py             │
│                               │                         │
│                            🔊 ~3s                        │
└──────────────────────────────────────────────────────────┘

Project Structure

voice-buddy/
├── .claude-plugin/          # Plugin manifest
│   ├── plugin.json
│   └── marketplace.json
├── hooks/hooks.json         # Hook declarations (auto-registered)
├── bin/voice-buddy          # CLI wrapper (auto-added to PATH)
├── agents/                  # 7 style-specific agent personas
│   ├── voice-buddy-cute-girl.md
│   ├── voice-buddy-elegant-lady.md
│   ├── voice-buddy-warm-boy.md
│   ├── voice-buddy-secretary.md
│   ├── voice-buddy-steward.md
│   ├── voice-buddy-cyber-girl.md
│   └── voice-buddy-kawaii.md
├── commands/voice-buddy.md  # /voice-buddy slash command
├── personas/                # Style definitions (TTS voice, rate, pitch)
├── templates/               # Per-style phrase templates
├── assets/audio/            # 84 pre-packaged MP3 files
├── voice_buddy/             # Python core
│   ├── main.py              # Hook entry point
│   ├── config.py            # Cross-platform user config
│   ├── styles.py            # Style definition loader
│   ├── context.py           # Event analysis
│   ├── response.py          # Template selection + audio ID
│   ├── tts.py               # edge-tts synthesis
│   ├── player.py            # Cross-platform audio playback
│   ├── injector.py          # Stop hook handler
│   ├── subagent_tts.py      # Agent TTS entry point
│   ├── generate_audio.py    # Dev utility: generate MP3 assets
│   └── cli.py               # CLI commands
└── tests/                   # unit + contract + release-gate suites

Development

git clone https://github.com/luyao618/voice-buddy.git
cd voice-buddy

# Editable install with dev extras, pinned to the verified versions
pip install -c constraints.txt -e ".[dev]"

# Run tests
python3 -m pytest tests/ -v

# Dependency advisory check
python3 -m pip_audit

# Regenerate pre-packaged audio (after editing templates)
python3 -m voice_buddy.generate_audio

Dependency files

File	Role
`pyproject.toml`	Source of truth. Declares what versions are allowed (ranges).
`constraints.txt`	Fully-pinned direct and transitive versions — what was actually verified.
`requirements.txt` / `requirements-dev.txt`	Convenience installers that mirror `pyproject.toml`.

Install reproducibly by passing the constraints file:

pip install -c constraints.txt .            # runtime only
pip install -c constraints.txt -e ".[dev]"  # development

Omitting -c constraints.txt still works and resolves anything inside the declared ranges — use that to pick up newer versions deliberately.

Updating dependencies:

Edit the range in pyproject.toml (and mirror it in requirements*.txt).
Run PYTHON=python3.10 bash scripts/regen-constraints.sh to re-resolve and re-pin. It needs only CPython's bundled venv/ensurepip — no uv, no preinstalled pip. Regenerate on the 3.10 floor, not a newer interpreter. aiohttp and pytest gate transitives behind python_version < "3.11" (async-timeout, exceptiongroup), so a resolve on 3.11+ silently omits them and the result installs everywhere except the floor. The script enforces this and refuses any other version; the extra pins are inert on newer interpreters, since a constraint only binds a package actually being installed. A failed resolve exits non-zero and leaves constraints.txt untouched.
Run the suite and python3 -m pip_audit.

tests/test_packaging.py fails if the three files disagree — on an extra dependency, a differing version range, a dropped environment marker, an unpinned constraint, or a pin that falls outside its declared range. It also walks the dependency closure, so removing any transitive pin from constraints.txt fails the suite rather than silently weakening the lock, and asserts the floor-gated pins are present so an off-floor regen is caught on any interpreter.

Supported Python versions

3.10 – 3.14, each verified by installing the constrained dependency set into a fresh virtualenv and running the full suite. requires-python = ">=3.10,<3.15" encodes that matrix on both ends, so anything outside it is rejected at resolve time rather than installing and failing later.

3.9 is not supported. pyobjc-core publishes no 3.9 wheel and its source build fails, and pytest 9.x / pip 26.x both declare `Requires-Python

=3.10`. The floor makes installers reject 3.9 up front rather than failing partway through a build.

3.15+ is not supported yet — it doesn't exist, so it can't have been tested, and pyobjc tracks the CPython C-ABI closely enough that a new minor is exactly where this package would break. To add one: append the minor to SUPPORTED_MINORS in tests/test_packaging.py, add its classifier, raise the cap, and run the matrix on it. The contract test fails if the cap and SUPPORTED_MINORS disagree, so the cap can't be raised without the declaration.

The packaging contracts run on every supported version, including the 3.10 floor: tomllib is 3.11+, so the dev extra carries a tomli backport for 3.10 rather than skipping the tests there.

中文

Voice Buddy 让 Claude Code 不再沉默。CC 是一个有性格的语音伙伴，在关键时刻用你选择的风格陪伴你编程。

打开 Claude Code  → "欢迎回来，今天也要加油哦~"         (cute-girl)
需要你注意时      → "Boss, your attention is needed"    (secretary)
任务完成          → "Senpai、バグ直したよ~"              (kawaii)
关闭会话          → "辛苦了，好好休息一下"               (warm-boy)

7 种风格，3 种语言

风格	语言	描述	TTS 语音	默认称呼
cute-girl	中文	可爱甜美	zh-CN-XiaoyiNeural	Master
elegant-lady	中文	优雅知性	zh-CN-XiaoxiaoNeural	Master
warm-boy	中文	温暖体贴	zh-CN-YunxiNeural	Master
secretary	英文	专业干练	en-US-JennyNeural	Boss
steward	英文	英式管家	en-GB-RyanNeural	Sir
cyber-girl	英文	赛博机器姬	en-GB-SoniaNeural	Commander
kawaii	日文	元气可爱	ja-JP-NanamiNeural	Senpai

工作原理

Voice Buddy 接入 Claude Code 的 Hook 系统，在关键事件触发时播放语音。

三层音频架构：

通道	延迟	适用场景	实现方式
Pre-packaged	<100ms	开始、结束	预生成 MP3 直接播放
实时 TTS	~1s	通知提醒	edge-tts 实时合成（含 nickname 替换）
AI 总结	~3s	任务完成	Claude subagent 生成 + TTS

安装

环境要求

Claude Code — 已在 v2.1.179 上验证
Python 3.10 – 3.14（不支持 3.9，原因见下方 Development 小节）
音频播放器（macOS: afplay 内置, Linux: paplay/aplay/mpg123）

Claude Code 兼容性说明. Voice Buddy 的 skill 位于 skills/voice-buddy/SKILL.md，hooks 使用 exec form（command + args）并通过 ${CLAUDE_PLUGIN_ROOT} 解析路径，两者都是当前的 plugin contract。旧的 commands/ 扁平布局与 shell-form hooks 仍可加载，但已不是官方默认写法。 exec form 与 ${CLAUDE_PLUGIN_ROOT} 替换已在 v2.1.179 实测通过；官方文档未给出这两项的最低版本要求，因此更早的版本属于「未测试」而非「已知不可用」。发布前的校验命令为 claude plugin validate . --strict。

从 Plugin Marketplace 安装

在 Claude Code 中依次运行以下两条命令：

/plugin marketplace add luyao618/voice-buddy   # 1. 添加插件源
/plugin install voice-buddy                                 # 2. 安装插件

安装时会提示选择风格和称呼，也可以直接回车使用默认值（cute-girl / Master）。

安装完成后 Hook 自动注册，无需手动配置。

升级

/plugin marketplace update voice-buddy-marketplace   # 刷新插件源
/plugin update voice-buddy@voice-buddy-marketplace   # 应用新版本

之后重启 Claude Code。旧版本遗留的监听器会被终止并替换，保证同时只有一个在跑。

必须用完整的 voice-buddy@voice-buddy-marketplace。只写插件名会报 Plugin "voice-buddy" not found。

卸载

/plugin uninstall voice-buddy

配置会特意保留，重新安装即可恢复你的风格与称呼。如果想一并删除：

# macOS
rm -rf ~/Library/Application\ Support/voice-buddy
# Linux
rm -rf ~/.config/voice-buddy

该目录只存放 config.json、调试日志和快捷键监听器的运行时状态，不会影响系统其他部分。

推荐：添加权限白名单

在 .claude/settings.json 的 allow 列表中加入以下规则，可以跳过每次语音播放的权限确认，体验更流畅：

"Bash(PYTHONPATH=* python3 -m voice_buddy*)",
"Bash(PYTHONPATH=* python3 -m voice_buddy.*)"

配置

安装后，在 Claude Code 对话中输入 /voice-buddy 即可进入配置面板。

也可以直接告诉 Claude 你想要的设置，例如：

"帮我把 Voice Buddy 切到 kawaii 风格"
"把称呼改成 Senpai"
"关掉 notification 的语音"

或者使用 CLI 命令（插件安装后 voice-buddy 命令自动可用）：

# 切换风格
voice-buddy config --style kawaii

# 修改称呼
voice-buddy config --nickname Senpai

# 同时修改
voice-buddy config --style secretary --nickname Boss

# 禁用/启用特定事件
voice-buddy config --disable notification
voice-buddy config --enable notification

# 全局开关
voice-buddy on
voice-buddy off

# 试听
voice-buddy test sessionstart
voice-buddy test notification

配置文件位置：

macOS: ~/Library/Application Support/voice-buddy/config.json
Linux: ~/.config/voice-buddy/config.json
Windows: %APPDATA%\voice-buddy\config.json

平台支持说明. macOS 与 Linux 有 CI 覆盖。Windows 有对应的配置路径、hook 也是按可运行编写的，但未经测试 —— 全局快捷键仅限 macOS，进程管理的部分实现依赖 POSIX 原语。Windows 请按 best-effort 对待。

支持的事件

事件	触发时机	音频来源
SessionStart	打开 Claude Code	Pre-packaged MP3
SessionEnd	关闭会话	Pre-packaged MP3
Notification	Claude 发送通知	实时 TTS（含称呼）
Stop	Claude 完成任务	AI 生成个性化总结

🔇 全局快捷键停止当前播放（macOS）

随时按 F2 可以立即让 Voice Buddy 安静下来——同事突然过来讲话、接电话、临时会议都不会被打扰。

首次安装后必做的设置

Marketplace 安装器只会复制插件文件，不会安装 Python 依赖，也不会授权 macOS 权限。安装后需要做下面三件事，做一次就行：

第 ① 步 — 安装 macOS 依赖

pip3 install 'pyobjc-framework-Quartz>=10.0,<13'

如果你电脑上有多个 Python 解释器（系统自带 / Homebrew / pyenv / conda / depot_tools 等），必须装到运行 voice-buddy 的那一个里。先用 doctor 查出来：

voice-buddy hotkey-doctor --non-interactive
# 看这一行：[OK] python interpreter   current=/path/to/python3

然后用那个路径直接装：

/path/to/python3 -m pip install 'pyobjc-framework-Quartz>=10.0,<13'

第 ② 步 — 把 F1/F2 设成标准功能键

打开 系统设置 → 键盘 → 键盘快捷键… → 功能键，把 「将 F1、F2 等键用作标准功能键」打勾。

不勾的话 F2 会被 macOS 截为「亮度增加」，应用根本收不到这个键。

第 ③ 步 — 给 Python 解释器授权辅助功能

跑 doctor 拿到精确路径：

voice-buddy hotkey-doctor --non-interactive
# 看这一行：[FAIL] Accessibility granted   grant Accessibility to: /path/to/python3

然后打开 系统设置 → 隐私与安全性 → 辅助功能：

点 + 号
弹出文件选择器后按 ⌘ + Shift + G，把上面 python 所在目录粘进去回车
如果 python3 是符号链接（depot_tools / pyenv shim 经常这样），无法直接选中。改选它指向的真实二进制文件（例如 python3.11）。检查方法：
```
ls -la /path/to/python3
```
如果文件选择器还是不让选，打开 Finder 把那个 python 二进制直接拖到「辅助功能」列表里：
```
open /path/to/bin
```
确认新增那一行右边的开关是 绿色（开）

第 ④ 步 — 重启 Claude Code

关闭并重新打开 Claude Code，让 SessionStart 钩子触发监听器进程启动。

验证

voice-buddy hotkey-doctor

期望看到（互动模式会让你按一次 F2）：

[OK  ] python interpreter
[OK  ] pyobjc importable
[OK  ] Accessibility granted  /usr/local/bin/python3 (EventTap created)
[OK  ] F-key fn-mode          F2 keydown observed
[OK  ] coord.lock writable
[OK  ] listener liveness      pid=...
[OK  ] version handshake
[OK  ] sessions registry
[OK  ] playback_pids sanity

任何一行 FAIL 或 WARN，detail 列会告诉你怎么修。

日常使用

# 不靠快捷键也能停（例如另一个终端窗口）
voice-buddy stop

# 重启快捷键监听器（按 PID 停止；下个 session 会自动重新拉起）
voice-buddy hotkey-restart

# 改键
voice-buddy config --hotkey F3

# 关闭 / 开启快捷键监听
voice-buddy config --disable-hotkey
voice-buddy config --enable-hotkey

# 体检
voice-buddy hotkey-doctor
voice-buddy hotkey-doctor --non-interactive   # CI 友好，跳过按键
voice-buddy hotkey-doctor --json              # 机器可读

监听器生命周期

监听器是单例进程，整台机器只有一个，不管开几个 Claude Code 窗口
启动：SessionStart 钩子检查是否有活监听器；没有就 spawn 一个（detached，约 500ms 冷启动）
退出：最后一个 Claude Code 会话关闭后，监听器内置的 30 秒空闲定时器会发现 sessions/ 空了，自动退出
F2 只打断当前正在播的那一段音频，不会取消还没开始播的 TTS 队列
完整 QA 清单见 docs/manual-tests.md

故障排查

现象	原因	解决
doctor 显示 `[FAIL] pyobjc importable`	pyobjc 没装到对的解释器	用第 ① 步对着 doctor 报告的路径重装
已经授权了，doctor 还报 `[FAIL] Accessibility`	授权了错的解释器，或路径是符号链接	重跑 doctor 看它给的路径；是符号链接就授权真实二进制
所有 OK 但 F2 还是不响应	监听器是在你装 pyobjc 之前 spawn 的，已经因为 import 失败退出	运行 `voice-buddy hotkey-restart`，然后开一个新 session
F2 突然失效，且 doctor 显示的监听器其实没在跑	监听器被强杀后 PID 被系统回收给了别的进程	`voice-buddy hotkey-restart` 会清掉这条陈旧记录，新 session 会重新拉起
升级 Python / 重建 venv 后 F2 失效	辅助功能授权绑定可执行文件路径，新 python 是不同文件	doctor 会显示 `[WARN] python interpreter` 含 `DRIFT`，重新授权新路径
想在 CI 里验证	—	`voice-buddy hotkey-doctor --non-interactive --json`

License

MIT

Name		Name	Last commit message	Last commit date
.claude-plugin		.claude-plugin
.github/workflows		.github/workflows
agents		agents
assets/audio		assets/audio
bin		bin
docs		docs
hooks		hooks
personas		personas
scripts		scripts
skills/voice-buddy		skills/voice-buddy
templates		templates
tests		tests
voice_buddy		voice_buddy
.gitignore		.gitignore
.pip-audit-ignore		.pip-audit-ignore
PLAN.md		PLAN.md
README.md		README.md
constraints.txt		constraints.txt
pyproject.toml		pyproject.toml
requirements-dev.txt		requirements-dev.txt
requirements.txt		requirements.txt

Latest commit

History

Folders and files

Repository files navigation