Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/zanezhao0708/voxagent/voice-controlnpx skills add zanezhao0708/voxagent --skill voice-controlgit clone --depth 1 https://github.com/zanezhao0708/voxagentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zanezhao0708/voxagent/voice-control)<a href="https://agentmods.dev/skills/zanezhao0708/voxagent/voice-control"><img src="https://agentmods.dev/badge/skills/zanezhao0708/voxagent/voice-control.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00067 | $0.00688 |
| Opus 5 | $0.00034 | $0.00344 |
| Sonnet 5 | $0.00013 | $0.00138 |
| Haiku 4.5 | $0.00007 | $0.00069 |
Grade A, and why
voice-control scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Voice Control
You have voice tools from the voxagent MCP server. Use them to interact by
voice — never guess what the user said, always call a tool.
When to use
- The user says "voice mode / 语音模式 / listen up" → switch to voice: all
further questions and confirmations go through
ask_user, summaries viaspeak. - The user asks you to "read this aloud / 念一下" →
speaka condensed version (max 2 sentences). - The user says "take dictation / 我说你们记" → call
listenand treat the transcript as their message.
Rules
- Spoken replies must be short. Max 2 sentences per
speakcall. Details stay in the chat; voice carries the headline. - Never invent audio input. If
listenreturns "[nothing heard]", say so and retry once, then fall back to text. - Language mirroring. Reply by voice in the language the user spoke. Transcripts may be mixed-language; follow the dominant one.
- Announce long tasks. Before a step that takes >30s (builds, big
refactors, test suites),
speaka one-line heads-up like "Tests are running, about a minute." - Errors are spoken too. If something fails while in voice mode,
speaka one-liner plus the fix you are attempting. - Degrade gracefully when voice fails. If
speakerrors or returns0, do not retry in a loop — deliver the same message as text and note that audio is unavailable. Voice is an enhancement, never a blocker.
Push-to-talk context
The user may drive you with the global push-to-talk hotkey instead of typing.
Transcripts then arrive as plain messages with no "voice mode" preamble. Treat
a short, imperative spoken-style message (e.g. "run the tests") as a voice
instruction and reply with a brief speak confirmation plus the normal text
detail.
When voice fails
listen→ "[nothing heard]": retry once, then continue in text.speakerrors: fall back to text; suggestvoxagent doctorto the user.- Repeated STT/TTS failures: tell the user to run
voxagent doctorand check mic permission /pip install 'voxagent[local]'.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 62 lines · 67 tokens per session scan A d4a382abc562
voice-control is a skill published in the GitHub repository zanezhao0708/voxagent (4 stars, last pushed 17d ago), licensed MIT. It adds 67 tokens to every session and 688 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
readme-benchmarks
Running the Kokoro-FastAPI benchmark + transcription-roundtrip suites and regenerating the README plots. Use when asked to run/refresh benchmarks, RTF/first-token plots, transcription sanity checks, or the long-form baseline.
api-contrib
Contributing to the Kokoro-FastAPI Python API: module layout, endpoint gating pattern, test expectations. Use when adding or changing endpoints, services, or inference code.
web-contrib
Contributing to the Kokoro-FastAPI web player: vanilla JS constraints, MSE/audio gotchas, unit and e2e test setup. Use when changing anything under web/.
integration-tests
Running the full e2e integration test suite (including Whisper roundtrip) against a live Kokoro server.
videoagent-audio-studio
Tired of juggling multiple audio APIs? This skill gives you one-command access to TTS, music generation, sound effects, and voice cloning. Use when you want to generate any audio without managing multiple API keys.
video-recap
从输入视频端到端生成中文解说成片。用户提供 .mp4 / .mov / .mkv / .webm，并要求添加旁白、 配音、总结、短剧/电视剧/电影/纪录片/科普解说时使用。负责编排 video- 技能链：视频理解 → Agent 制定故事与视听方案 → 剪辑 → 配音 → 合成。触发词：视频解说、视频旁白、生成解说、 视频 recap、video recap、voiceover、narration、auto-dub、recap。.