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 yoav0gal/agent-voice --skill create-speech-recordinggit clone --depth 1 https://github.com/yoav0gal/agent-voiceWrote 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/yoav0gal/agent-voice/create-speech-recording)<a href="https://agentmods.dev/skills/yoav0gal/agent-voice/create-speech-recording"><img src="https://agentmods.dev/badge/skills/yoav0gal/agent-voice/create-speech-recording/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/yoav0gal/agent-voice/create-speech-recording"><img src="https://agentmods.dev/badge/skills/yoav0gal/agent-voice/create-speech-recording.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 | $0.00024 | $0.00589 |
| Opus 5.5 | $0.00010 | $0.00236 |
| Sonnet 5.5 | $0.00005 | $0.00118 |
| Haiku 4.5 | $0.00002 | $0.00059 |
Grade A, and why
create-speech-recording scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 21d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create Speech Recording
Agent Voice works best in English. For another supported language, choose a
matching voice with agent-voice voices and pass --voice and --lang.
Prepare
Set RESPONSE_AS_TEXT to the text's spoken form.
-
Use speech-ready plain text, not serialized text. Convert literal
\n,\r,\t, , , and to normal spacing. To discuss one intentionally, write its spoken form, such as "backslash n". -
For supplied text, preserve every word and punctuation mark in order while translating presentation syntax into speech.
-
For a requested summary or explanation, write natural speech with the same meaning.
-
For tables, state the headers once and read each row as labeled values.
Record
agent-voice speak "$RESPONSE_AS_TEXT"
For long text, create a unique file in the system temporary directory. The operating system handles cleanup.
agent-voice speak --label "$LABEL" < "$RESPONSE_AS_TEXT_FILE"
Set LABEL to a short subject. For an exact requested filename, replace
--label "$LABEL" with --output "$PATH".
Use the returned path to deliver the recording. For speaker playback, add
-p; continue after the result reports playback.state: "started". Add
--play-after SECONDS to schedule it without blocking.
Deliver
Use default.md immediately after speak
returns a receipt with generation.state: "started"
Remote delivery
Remote delivery requires Agent Voice 0.10.0 or newer. Run agent-voice update
if an installed version does not recognize --viewer-base-url.
When delivery.browser_url uses HTTPS, or generation runs on another machine,
return the HTTPS player URL as [Listen](URL). This takes precedence over the
local delivery examples. Local file paths and desktop controls refer to the generating
machine; omit them from remote delivery. Do not use -p to play on that machine
unless the user asks. A localhost URL needs a configured private proxy and
agent-voice config --viewer-base-url HTTPS_ORIGIN before remote delivery works.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 21d ago Changed · +16 lines 83ed4995cdea
- 1mo ago First seen · 59 lines · 24 tokens per session scan A 886ac2b7c501
create-speech-recording is a skill published in the GitHub repository yoav0gal/agent-voice (11 stars, last pushed 12d ago), licensed MIT. It adds 24 tokens to every session and 589 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
kanban-tasks
A task-board workflow for breaking work into cards and moving them through stages such as inbox, ready, doing, review, and done. It also includes a background service and a web interface for the board.
context-recovery
Recovers project handoff context from local Codex, Claude Code, Gemini, CodeBuddy, and codexmate-derived sessions. Use when the user asks what happened in prior project/PR/branch/file/error work, needs a handoff brief, wants old decisions or validations recovered, or asks to summarize cross-session project activity…
ocr
OCR image files using the Qwen VL model via the bundled scripts/ocr.py (lives in this skill's directory, NOT your cwd). Use when the user asks to extract text from an image, perform OCR on a photo or screenshot, or recognize characters in an image file (.jpg, .jpeg, .png, .gif, .webp). Requires QWENAPIKEY and…
daily-digest
Build a daily digest from curated sources. Follow the output contract strictly.
duckdb-analyst
Use for any analytical question over ./data/.parquet. Prefer DuckDB, always show the SQL.
onboard
Link this project to a Pad workspace and set it up.