openrouter-tts — Claude Code skill: what it does, how to install
openrouter-tts synthesizes speech audio from text using OpenRouter's speech API. Generates voiceovers, audiobook snippets, and audio files right from your prompt.
no license 327 installs ★ 327 updated 2026-10-24
npx skills add OpenRouterTeam/skills --skill openrouter-tts -a claude-codeWhat openrouter-tts does
The skill equips the assistant to call OpenRouter's POST /api/v1/audio/speech endpoint using curl or OpenAI-compatible client libraries. Successful calls return raw binary audio rather than a JSON payload, so the agent writes the bytes directly to a media file.
An OPENROUTER_API_KEY environment variable is mandatory. The default format returned by the endpoint is uncompressed pcm, which will fail to play if saved under a .mp3 extension. Setting response_format to mp3 explicitly is required for standard media players.
Voice names are provider-specific. OpenAI models accept identifiers like alloy or nova, whereas engines like Voxtral or Kokoro require their own custom tags. The skill also handles model discovery by querying the models endpoint with speech modality filters.
What it helps with
- Convert article drafts or scripts into playable mp3 audio
- Query OpenRouter for speech models and their supported voice names
- Adjust voice tone and pacing via OpenAI provider options
- Export raw pcm audio streams for custom processing pipelines
Example prompts for the agent
- Convert script.txt into an mp3 file using openai/gpt-4o-mini-tts-2025-12-15 with the alloy voice
- List all TTS models and supported voices available on OpenRouter
- Generate an audio clip with a slow, warm tone using provider instructions
How to install openrouter-tts
1Skills CLI (Vercel)
npx skills add OpenRouterTeam/skills --skill openrouter-tts -a claude-code2Manually: put the files in place
~/.claude/skills/openrouter-tts/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/openrouter-tts/
1Skills CLI (Vercel)
npx skills add OpenRouterTeam/skills --skill openrouter-tts -a codex2Manually: put the files in place
.agents/skills/openrouter-tts/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add OpenRouterTeam/skills --skill openrouter-tts -a cursor2Manually: put the files in place
.cursor/skills/openrouter-tts/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install OpenRouterTeam/skills skills/openrouter-tts2Manually: put the files in place
.github/skills/openrouter-tts/SKILL.md1Skills CLI (Vercel)
npx skills add OpenRouterTeam/skills --skill openrouter-tts -a gemini-cli2Platform CLI
gemini skills install https://github.com/OpenRouterTeam/skills --path skills/openrouter-ttsManually: put the files in place
.devin/skills/openrouter-tts/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add OpenRouterTeam/skills --skill openrouter-tts -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about openrouter-tts
How to install openrouter-tts in Claude Code?
Run npx skills add OpenRouterTeam/skills --skill openrouter-tts -a claude-code. Alternatively, copy the definition directly into ~/.claude/skills/openrouter-tts/SKILL.md.
What does openrouter-tts do?
It allows your coding assistant to synthesize speech from text files or prompts by routing requests to speech models hosted on OpenRouter.
Why is the generated mp3 file unplayable?
The endpoint outputs raw pcm bytes unless you set response_format to mp3. Saving raw samples with an .mp3 extension corrupts playback.