Narration and text-to-speech on the user's machine through Guaardvark's Audio Foundry (Chatterbox, Kokoro, Piper) and consent-gated voice cloning from a reference clip.
Install the "voice" agent skill from https://github.com/guaardvark/guaardvark/tree/main/.agents/skills/voice into .claude/skills/voice/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "voice", then confirm the skill loads.
Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
skills CLI
$ npx skills add guaardvark/guaardvark --skill voice -a codex
Project install goes to .agents/skills/; add -g for ~/.codex/skills/.
Install the "voice" agent skill from https://github.com/guaardvark/guaardvark/tree/main/.agents/skills/voice into .agents/skills/voice/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "voice", then confirm the skill loads.
Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add guaardvark/guaardvark --skill voice -a cursor
Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.
Install the "voice" agent skill from https://github.com/guaardvark/guaardvark/tree/main/.agents/skills/voice into .cursor/skills/voice/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "voice", then confirm the skill loads.
Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
skills CLI
$ npx skills add guaardvark/guaardvark --skill voice -a gemini-cli
Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.
Install the "voice" agent skill from https://github.com/guaardvark/guaardvark/tree/main/.agents/skills/voice into .gemini/skills/voice/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "voice", then confirm the skill loads.
Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
GitHub CLI
$ gh skill install guaardvark/guaardvark voice
Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
skills CLI
$ npx skills add guaardvark/guaardvark --skill voice -a github-copilot
Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.
Install the "voice" agent skill from https://github.com/guaardvark/guaardvark/tree/main/.agents/skills/voice into .github/skills/voice/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "voice", then confirm the skill loads.
GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
skills CLI
$ npx skills add guaardvark/guaardvark --skill voice -a opencode
OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
Install the "voice" agent skill from https://github.com/guaardvark/guaardvark/tree/main/.agents/skills/voice into .opencode/skills/voice/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "voice", then confirm the skill loads.
OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
Skill name
voice
GitHub stars
257
Token cost
~798 tokens
SKILL.md length
289 words
Files
1
Skills in repo
15
Repo updated
First seen
Licence
MIT
At a glance
Narration and text-to-speech on the user's machine through Guaardvark's Audio Foundry (Chatterbox, Kokoro, Piper) and consent-gated voice cloning from a reference clip.
Works in 3 steps: The reference must go through the upload… → Generate with "backend": "chatterbox",… → Before uploading, ask whether the voice…
The user wants a voiceover
SKILL.md covers Which engine, Speak a line or a script, Clone a voice (consent-gated) and Rules
Calls curl
What it does
Voice is an agent skill from guaardvark/guaardvark. Narration and text-to-speech on the user's machine through Guaardvark's Audio Foundry (Chatterbox, Kokoro, Piper) and consent-gated voice cloning from a reference clip. Use when the user wants a voiceover, narration of a script, a spoken line, or "make it sound like this voice".
Its SKILL.md is about 800 tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Media & Creative, covering Text to speech and voice. It works with Model Context Protocol. The repository describes itself as: Self-hosted AI studio on your own GPU: chat with your files (RAG), screen and browser agents, coding swarms, LoRA training, an MCP server for Claude Code and Cursor, and local… The licence is MIT.
When your agent uses it
The user wants a voiceover
Narration of a script
Make it sound like this voice
Example prompts
“s machine through Guaardvark”
“make it sound like this voice”
“/voice”
Workflow steps
3 steps, taken from the first numbered list in SKILL.md.
1The reference must go through the upload route; that is what records consent. Arbitrary
2Generate with "backend": "chatterbox", "reference_clip_path": "".
3Before uploading, ask whether the voice belongs to the user or someone who consented. Do not
What it can do on your machine
Read from SKILL.md and the folder at commit 2a33110. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
curl
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use curl, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Voice loads about 798 tokens when it runs. Until then it costs about 71 tokens; SKILL.md has 289 words of instructions outside code blocks.
Always· name and description, kept in context so the agent knows when to use it
~71
When it runs· the whole SKILL.md, loaded when a task matches
~798
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
Download SKILL.mdSave it as .claude/skills/voice/SKILL.md (or your agent's skills folder).
name
voice
description
Narration and text-to-speech on the user's machine through Guaardvark's Audio Foundry (Chatterbox, Kokoro, Piper) and consent-gated voice cloning from a reference clip. Use when the user wants a voiceover, narration of a script, a spoken line, or "make it sound like this voice".
Voice with Guaardvark
Read setup first. Expressive voices need the audio_foundry plugin running;
Piper works without it. B=${GUAARDVARK_URL:-http://localhost:5000}.
GET $B/api/audio-foundry/voices lists what is installed. GET $B/api/voice/voices lists Piper voices.
Speak a line or a script
Over MCP, call generate_speech (text up to 3000 characters, optional voice such as
af_heart, engine auto | kokoro | chatterbox); it waits and returns the file with a download
link. Cloning is not offered over MCP. Without MCP, or for the Chatterbox knobs, use REST:
bash
curl -s -X POST $B/api/audio-foundry/generate/voice -H 'Content-Type: application/json' -d '{
"text": "The line to speak.",
"backend": "auto", # auto | chatterbox | kokoro
"voice_id": "af_heart", # Kokoro voice, or omit
"emotion": "calm", # Chatterbox preset, or omit
"exaggeration": 0.5, "cfg_weight": 0.5, "temperature": 0.8, # Chatterbox knobs, optional
"seed": 7, "output_format": "wav", "async": true
}'
Short text returns the file directly (path, document_id). With "async": true or long
text you get 202 {"job_id"}: poll GET $B/api/audio-foundry/jobs/<job_id> until status
is done; the result has path and document_id. Cancel: POST .../jobs/<job_id>/cancel.
Multi-section narration with pauses: POST $B/api/voice/narrate{"script": "...", "engine": "kokoro", "voice": "...", "pause_between_sections": 0.6, "output_format": "wav"}.
Piper only: POST $B/api/voice/text-to-speech {"text", "voice": "libritts"} returns audio_url.
Clone a voice (consent-gated)
The reference must go through the upload route; that is what records consent. Arbitrary
file paths are refused with 403.bash
curl -s -X POST $B/api/audio-foundry/voice-clips/upload -F file=@/abs/path/ref.wav -F name="Narrator sample"
The response gives the stored path. GET $B/api/audio-foundry/voice-clips lists clips.
Generate with "backend": "chatterbox", "reference_clip_path": "<that path>".
Before uploading, ask whether the voice belongs to the user or someone who consented. Do not
clone a public figure or anyone who has not agreed. Refuse politely if unclear.
Rules
10 to 20 seconds of clean speech is enough for a clone; more is not better.
Say which engine ran (the response reports it); auto falls back to Kokoro on a Chatterbox error.
Audio files are local under data/outputs/; they also appear in the Audio library page.
Voice next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
Transcribe a video with Sonilo and translate the transcript into editable .srt files — one per target language, plus the detected source language — so the wording can be read and corrected before…
Build consistent characters, environments and props in Guaardvark's Cast Library and train LoRAs for them locally (reference photos → vision bible → sample plan → approved samples → training).
Generate or edit images on the user's own GPU through Guaardvark: single images, instruction edits, background cut-outs, inpaint and outpaint, consistent characters from the Cast Library, and batch…
Generate full songs with vocals or instrumentals (ACE-Step) and sound effects or ambience (Stable Audio Open) on the user's GPU through Guaardvark's Audio Foundry.
Operate a running Guaardvark: GPU and VRAM state, plugin start/stop, logs, Celery tasks, the Interconnector sync to other machines, overnight RAG autoresearch, and infographics.
Launch and watch Guaardvark's Swarm Orchestrator: parallel coding agents, each in its own git worktree, working a markdown plan and merging back deterministically.
Narration and text-to-speech on the user's machine through Guaardvark's Audio Foundry (Chatterbox, Kokoro, Piper) and consent-gated voice cloning from a reference clip. Voice is an agent skill from guaardvark/guaardvark. Narration and text-to-speech on the user's machine through Guaardvark's Audio Foundry (Chatterbox, Kokoro, Piper) and consent-gated voice cloning from a reference clip.
When should I use Voice?
Voice fits situations like: the user wants a voiceover; narration of a script; make it sound like this voice.
How do I install Voice in Claude Code?
Run `npx skills add guaardvark/guaardvark --skill voice -a claude-code`. Or copy the skill folder (.agents/skills/voice in guaardvark/guaardvark) into .claude/skills/voice in your project. Claude Code loads it when a task matches its description.
How do I install Voice in Codex?
Run `npx skills add guaardvark/guaardvark --skill voice -a codex`. Or copy the skill folder (.agents/skills/voice in guaardvark/guaardvark) into .agents/skills/voice in your project. Codex loads it when a task matches its description.
Can I use Voice in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add guaardvark/guaardvark --skill voice -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/voice, .gemini/skills/voice, .github/skills/voice and .opencode/skills/voice in your project.
What does Voice need to run?
Going by SKILL.md and its folder, Voice needs the command-line tools its instructions call (curl).
Does Voice access the network?
SKILL.md contains no URLs. Its commands use curl, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Voice safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Voice use?
Voice is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Voice use?
About 798 tokens (SKILL.md is roughly 3.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Voice?
Skills that share tags, products or a category with Voice: Audio Ducking (sonilo-ai/skills, 115 stars), Auto Dubbing (sonilo-ai/skills, 115 stars), Proofread (sonilo-ai/skills, 115 stars) and Task Recovery (sonilo-ai/skills, 115 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Voice?
guaardvark (a GitHub user) maintains it in guaardvark/guaardvark, which has 257 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on October 9, 2026.
Source: guaardvark/guaardvark on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.
