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 commands/josephscript/claude-speak/speak-voicegit clone --depth 1 https://github.com/JosephScript/claude-speakWrote 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/commands/josephscript/claude-speak/speak-voice)<a href="https://agentmods.dev/commands/josephscript/claude-speak/speak-voice"><img src="https://agentmods.dev/badge/commands/josephscript/claude-speak/speak-voice.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 | $0.00029 | $0.00405 |
| Opus 5 | $0.00015 | $0.00202 |
| Sonnet 5 | $0.00006 | $0.00081 |
| Haiku 4.5 | $0.00003 | $0.00040 |
Grade B, and why
speak-voice scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **No argument**: report the current voice. Run `cat ~/.claude/claude-speak/voice 2>/dev/null`; if empty/missing, the voice is "system default". What it actually says
Based on the argument passed to this command, do exactly one of the following via Bash, then report the result in one short sentence — nothing else besides that (for list, show the list itself, not a one-line summary):
- No argument: report the current voice. Run
cat ~/.claude/claude-speak/voice 2>/dev/null; if empty/missing, the voice is "system default". list: run"${CLAUDE_PLUGIN_ROOT}/scripts/tts-backend.sh" list-voicesand show the result — these are the installed English voices available to pick from on this machine's backend (macOSsayor Linux espeak-ng).reset: clear back to the system default voice:rm -f ~/.claude/claude-speak/voice.- Any other value: treat it as a voice name. First verify it exists:
"${CLAUDE_PLUGIN_ROOT}/scripts/tts-backend.sh" validate-voice "<name>". If that fails, tell the user the voice name wasn't recognized and suggest running/speak-voice list. If it succeeds, write it:mkdir -p ~/.claude/claude-speak && echo "<name>" > ~/.claude/claude-speak/voice.
This voice applies everywhere speech is used: the automatic /toggle-speak Stop hook, /speak (local mode), and /speak llm. Voice names are backend-specific (macOS uses names like Daniel/Samantha, espeak-ng uses codes like en-us/en-gb) — a voice set on one machine's backend won't mean anything on another.
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 · 14 lines · 29 tokens per session scan B b674ecf6041b
speak-voice is a command published in the GitHub repository JosephScript/claude-speak (3 stars, last pushed 1mo ago), licensed MIT. It adds 29 tokens to every session and 405 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.