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 skills add Tomorrow-You/voice-bridge --skill speakgit clone --depth 1 https://github.com/Tomorrow-You/voice-bridgeWrote 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/skills/tomorrow-you/voice-bridge/speak)<a href="https://agentmods.dev/skills/tomorrow-you/voice-bridge/speak"><img src="https://agentmods.dev/badge/skills/tomorrow-you/voice-bridge/speak/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tomorrow-you/voice-bridge/speak"><img src="https://agentmods.dev/badge/skills/tomorrow-you/voice-bridge/speak.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00034 | $0.00398 |
| Opus 5 | $0.00017 | $0.00199 |
| Sonnet 5 | $0.00007 | $0.00080 |
| Haiku 4.5 | $0.00003 | $0.00040 |
Grade A, and why
speak 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 9d 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.
What it actually says
Voice Bridge -- Speak Aloud
When the user's message starts with "speak" (case-insensitive), wrap your entire response in <speak>...</speak> tags.
Rules
- ONLY use
<speak>tags when the user explicitly starts their message with "speak" - Strip the "speak" keyword before processing the actual request
- Inside
<speak>tags, write naturally for listening:- No markdown formatting (no
**bold**,# headers,- lists) - No code blocks or inline code
- No file paths or URLs
- No tables
- Spell out abbreviations on first use
- No markdown formatting (no
- Keep spoken responses concise -- under 2000 characters
- If the user just says "speak" with no follow-up, speak a brief status or greeting
Examples
User: "speak what does this function do?"
Response: <speak>This function takes a user ID and returns their profile data from the database. It first checks the cache, and if there's a miss, it queries the users table and caches the result for five minutes.</speak>
User: "speak summarize the changes"
Response: <speak>The main changes are: first, I added a new authentication middleware that checks JWT tokens on every request. Second, I updated the user model to include an email verification field. And third, I added rate limiting to the login endpoint to prevent brute force attacks.</speak>
User: "what does this function do?" (no "speak" prefix)
Response: Normal markdown response, NO <speak> tags.
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.
- 9d ago First seen · 34 lines · 34 tokens per session scan A c5891bd795f5
speak is a skill published in the GitHub repository Tomorrow-You/voice-bridge (0 stars, last pushed 4mo ago), licensed MIT. It adds 34 tokens to every session and 398 once invoked, about $0.0002 per session on Opus 5. 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.
Other skills, from other repositories
talkback
Speak an explanation out loud while working in any project — tiered text-to-speech with a pluggable backend (ElevenLabs by default and quota-guarded, macOS say via --fast for free instant local speech, local OmniVoice as an unlimited private tier). Markdown-aware, so code fences, URLs and deep paths collapse to short…
dialogue
Create multi-speaker dialogue. Write conversations between characters with different voices.
speak
Speak text aloud with voice selection, emotion, and effects. Say anything naturally.
status
Check voice engine health and configuration.
voices
List available voices and presets. Browse the full voice roster.
speak
Control text-to-speech for Claude Code responses. Turn on/off, adjust voice, speed, sentence count, and text cleaning.