elevenlabs-local-dev-loop
Use when setting up a local ElevenLabs dev environment for a TTS/voice project and you need SDK mocking, hot reload, quota-aware iteration, and audio-output testing that does not burn character quota during development. Trigger with "elevenlabs dev setup", "elevenlabs local development", "elevenlabs dev environment", "develop with elevenlabs", "test elevenlabs locally".
Allowed Tools
Provided by Plugin
elevenlabs-pack
Claude Code skill pack for ElevenLabs (18 skills)
Installation
This skill is included in the elevenlabs-pack plugin:
/plugin install elevenlabs-pack@claude-code-plugins-plus
Click to copy
Instructions
ElevenLabs Local Dev Loop
Overview
Set up a fast, cost-effective local development workflow for ElevenLabs audio projects. The loop centers on three moves — mock the SDK so unit tests never burn character quota, gate real API calls behind an explicit ELEVENLABS_INTEGRATION=1 flag, and select a cheaper model in dev while keeping the high-quality model for production — with tsx watch hot reload and a quota checker to round out the cycle.
Follow the high-level flow below to scaffold the project, then drill into references/implementation.md for the full code of every step and references/examples.md for worked end-to-end runs.
Prerequisites
Before starting, confirm your environment is ready:
- The
elevenlabs-install-authsetup is complete, so the SDK (@elevenlabs/elevenlabs-js) is installed andELEVENLABS_API_KEYis available in.env.local. - Node.js 18+ with
npmorpnpm. vitestinstalled as the test runner (recommended) — it powers the mock layer and the integration-test guard.
Instructions
Work through the six steps in order. Each is summarized here; the full code for every step lives in references/implementation.md.
- Project structure — lay out
src/elevenlabs/(client, config, tts),tests/__mocks__/andtests/fixtures/sample.mp3, a git-ignoredoutput/, and.env.local/.env.example. Full tree in the reference. - Environment configuration — write an environment-aware
config.tsthat picks the model and output format byNODE_ENV. This is the essential skeleton:
// src/elevenlabs/config.ts
export function loadConfig() {
const env = process.env.NODE_ENV || "development";
return {
apiKey: process.env.ELEVENLABS_API_KEY || "",
// cheaper/faster in dev, best quality in prod
modelId: env === "production"
? "eleven_multilingual_v2" // 1.0 credits/char
: "eleven_flash_v2_5", // 0.5 credits/char, ~75ms
defaultVoiceId: process.env.ELEVENLABS_VOICE_ID || "21m00Tcm4TlvDq8ikWAM",
outputFormat: "mp3_22050_32", // smaller files for dev
};
}
- Mock the SDK — write
tests/__mocks__/elevenlabs.tsthat returns thesample.mp3fixture fromtextToSpeech.convert/streamand stubsvoices.getAllanduser.get, so unit tests cost nothing. - Development scripts — add
dev(tsx watch),test,test:watch,test:integration,generate, andquotascripts topackage.json. - Quota-aware development — add
src/check-quota.tsthat readsuser.subscriptionand exits non-zero when fewer than 1000 characters remain, so a low balance fails fast. - Integration test guard — write
tests/tts.test.tswhere the real-API test isit.skipIf(!useRealApi)and only runs underELEVENLABS_INTEGRATION=1; the mocked test always runs.
See references/implementation.md for the complete, copy-pasteable code for each step.
Output
- Working development environment with hot reload via
tsx watch - Mock layer that avoids API calls and character charges during dev
- Quota checker to prevent surprise billing
- Integration test guard pattern (
ELEVENLABS_INTEGRATION=1) - Environment-aware model selection (cheap in dev, quality in prod)
Error Handling
| Error | Cause | Solution |
|---|---|---|
MODULE_NOT_FOUND |
SDK not installed | npm install @elevenlabs/elevenlabs-js |
| Mock returns undefined | Mock not wired | Check vi.mock path matches import |
| Integration test fails | No API key | Set ELEVENLABS_API_KEY in .env.local |
| Quota exceeded in dev | Running real API calls | Use mock layer; run npm run quota first |
Examples
Four worked runs of the loop — full walkthroughs in references/examples.md:
- Zero-cost unit tests —
npm run testdrives the service through the mock client, passes offline, and never touches the API or your quota. - Quota preflight —
npm run quotaprintsCharacters: 500 / 10,000 (5.0% used)and exits1when fewer than 1000 characters remain, blocking a paid run before it starts. - Opt-in integration run —
npm run test:integrationsetsELEVENLABS_INTEGRATION=1, flipping theit.skipIf(!useRealApi)test on so the real API is hit only when you ask for it. - Hot-reload iteration —
npm run dev(tsx watch) restarts on save; with the dev model (eleven_flash_v2_5) and mocks, each loop stays fast and free.
Resources
Next Steps
Once the dev loop is running, move on to production-ready code: see the elevenlabs-sdk-patterns skill for streaming, retries, and voice-management patterns you can layer on top of this environment.