voice-test

voice-test is a command for Claude Code from jamditis/audiobash. It costs 0 tokens per session (2,199 once invoked), scanned A, original, MIT.

A command for checking an AudioBash voice-transcription system from recorded audio through provider selection, API keys, and agent-mode prompts.

In plain words
What is it for?
It helps test Gemini models and, when implemented, OpenAI Whisper and Deepgram, including their settings, audio format support, error handling, and key storage.
Why use it?
It helps reveal whether transcription providers, credentials, audio handling, errors, and cost calculations work together.

Command for Claude Code

Install

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.

agentmods
npx agentmods add commands/jamditis/audiobash/voice-test
Clone the repo
git clone --depth 1 https://github.com/jamditis/audiobash

Made for: Claude Code.

Wrote 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.

agentmods badge for voice-test

README.md
[![agentmods](https://agentmods.dev/badge/commands/jamditis/audiobash/voice-test.svg)](https://agentmods.dev/commands/jamditis/audiobash/voice-test)
Your own site
<a href="https://agentmods.dev/commands/jamditis/audiobash/voice-test"><img src="https://agentmods.dev/badge/commands/jamditis/audiobash/voice-test.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,199 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.02199
Opus 5 $0.00000 $0.01099
Sonnet 5 $0.00000 $0.00440
Haiku 4.5 $0.00000 $0.00220

Measured 3d ago against content hash 15382398e3e3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

voice-test 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.

.claude/commands/voice-test.md · 335 lines

How it starts

The opening of the file, as written. The whole thing — 335 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Voice Pipeline Test

Test the complete AudioBash voice transcription pipeline including all providers, API key configuration, and agent mode prompts.

Your Role

You are a QA engineer testing the voice transcription system end-to-end.

Test Categories

1. Transcription Provider Verification

AudioBash supports multiple transcription providers. Verify each is correctly implemented:

Providers to check:

  • Gemini 2.0 Flash - Google's experimental model
  • Gemini 1.5 Flash - Google's production model
  • Gemini 1.5 Pro - Google's advanced model
  • OpenAI Whisper - If implemented
  • Deepgram - If implemented

Files to review:

# Find transcription service implementation
find src/ -name "*transcription*" -o -name "*voice*"

# Check for provider definitions
grep -rn "gemini-2.0-flash\|gemini-1.5-flash\|whisper\|deepgram" src/ --include="*.ts" --include="*.tsx"

Verify each provider:

  • ✓ Model ID correctly specified
  • ✓ API endpoint configured
  • ✓ Audio format supported (WebM/Opus)
  • ✓ Error handling implemented
  • ✓ Cost calculation (if applicable)

2. API Key Configuration

Test API key storage and retrieval:

# Check safeStorage implementation
grep -rn "safeStorage\|getApiKey\|setApiKey" electron/main.cjs -A 5

# Check IPC handlers for API keys
grep -rn "ipcMain.handle.*api.*key" electron/main.cjs -B 2 -A 10

Expected behavior:

// GOOD: Secure storage
const key = await safeStorage.getPassword('gemini-api-key');

// BAD: Plaintext storage
const key = fs.readFileSync('api-key.txt');

Test scenarios:

  1. First launch - No API key stored

    • Should prompt user for API key
    • Should validate key format
    • Should store securely in safeStorage
  2. Key already configured

    • Should load from safeStorage
    • Should not prompt again
    • Should allow updating in settings
  3. Invalid key

    • Should show error message
    • Should not accept empty keys
    • Should allow retry

Read the full file on GitHub · 335 lines

Changes

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.

  1. 3d ago First seen · 335 lines · 0 tokens per session scan A 15382398e3e3

Subscribe to this mod's changes

voice-test is a command published in the GitHub repository jamditis/audiobash (5 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,199 tokens. 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.