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.
git clone --depth 1 https://github.com/Pr1m4lc0d3/wheel-to-talkWrote 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/pr1m4lc0d3/wheel-to-talk/wheel-to-talk-status)<a href="https://agentmods.dev/commands/pr1m4lc0d3/wheel-to-talk/wheel-to-talk-status"><img src="https://agentmods.dev/badge/commands/pr1m4lc0d3/wheel-to-talk/wheel-to-talk-status/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/commands/pr1m4lc0d3/wheel-to-talk/wheel-to-talk-status"><img src="https://agentmods.dev/badge/commands/pr1m4lc0d3/wheel-to-talk/wheel-to-talk-status.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 | $0.00009 | $0.00248 |
| Opus 5 | $0.00005 | $0.00124 |
| Sonnet 5 | $0.00002 | $0.00050 |
| Haiku 4.5 | $0.00001 | $0.00025 |
Grade B, and why
wheel-to-talk-status 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 12d 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.
2. Are voice settings correct? Read `~/.claude/settings.json` and report the What it actually says
Report the state of Wheel to Talk. Gather these facts, then summarize them plainly. Do not fix anything unless asked.
-
Is the backend running?
powershell -NoProfile -Command "Get-Process AutoHotkey64 -ErrorAction SilentlyContinue | Select-Object Id,ProcessName"More than one instance is a bug — two instances double every keystroke.
-
Are voice settings correct? Read
~/.claude/settings.jsonand report thevoiceblock.enabledmust be true. Note whetherautoSubmitis on, since that decides whether releasing the wheel sends the prompt or just fills it. -
What has the backend actually seen? Read the tail of the log:
%TEMP%\wheel-to-talk.logEach press logs
press -> space down, each releaserelease -> space up. If the log shows no presses when the user clicks, the hook is not firing — the click never reached the backend. If it shows presses but nothing records, the key is reaching Claude Code but not registering as held.
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.
- 12d ago First seen · 25 lines · 9 tokens per session scan B 5d4120a1165c
wheel-to-talk-status is a command published in the GitHub repository Pr1m4lc0d3/wheel-to-talk (1 stars, last pushed 14d ago), licensed MIT. It adds 9 tokens to every session and 248 once invoked, about $0.0000 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.
Other commands, from other repositories
start-talking
Start talking with Cadence Code.
jump-in
Jump In and add spoken guidance.
voice-settings
Choose Cadence Code speech models.
wrap-up
End the active Cadence Code conversation.
stts
User speaks the prompt, which is sent to the Model, the received response is spoken/read aloud in a loop.
prototype
You are building a proof-of-concept for the current Grainulator sprint. Read CLAUDE.md for sprint context and claims.json for existing research claims.