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 mydisha/keirouter --skill keirouter-sttgit clone --depth 1 https://github.com/mydisha/keirouterWrote 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/mydisha/keirouter/keirouter-stt)<a href="https://agentmods.dev/skills/mydisha/keirouter/keirouter-stt"><img src="https://agentmods.dev/badge/skills/mydisha/keirouter/keirouter-stt/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/mydisha/keirouter/keirouter-stt"><img src="https://agentmods.dev/badge/skills/mydisha/keirouter/keirouter-stt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 8 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00058 | $0.00783 |
| Opus 5.5 | $0.00023 | $0.00313 |
| Sonnet 5.5 | $0.00012 | $0.00157 |
| Haiku 4.5 | $0.00006 | $0.00078 |
Grade A, and why
keirouter-stt scanned grade A 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl $KEIROUTER_URL/v1/models/stt | jq '.data[].id' How it starts
The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
KeiRouter — Speech-to-Text
Requires KEIROUTER_URL (and KEIROUTER_KEY if auth enabled). See https://raw.githubusercontent.com/mydisha/keirouter/main/skills/keirouter/SKILL.md for setup.
Discover
curl $KEIROUTER_URL/v1/models/stt | jq '.data[].id'
curl "$KEIROUTER_URL/v1/models/info?id=openai/whisper-1"
model = provider/model, e.g. openai/whisper-1 or groq/whisper-large-v3-turbo.
Endpoint
POST $KEIROUTER_URL/v1/audio/transcriptions (OpenAI Whisper compatible, multipart/form-data)
| Field | Required | Notes |
|---|---|---|
model |
yes | from /v1/models/stt |
file |
yes | audio file (mp3, wav, m4a, webm, ogg, flac) |
language |
no | ISO-639-1 (e.g. en, id) |
prompt |
no | hint text to guide transcription |
response_format |
no | json (default) or text |
temperature |
no | 0–1 |
Examples
curl -X POST "$KEIROUTER_URL/v1/audio/transcriptions" \
-H "Authorization: Bearer $KEIROUTER_KEY" \
-F "model=openai/whisper-1" \
-F "[email protected]" \
-F "language=en"
JS (Node):
import { readFile } from "node:fs/promises";
const form = new FormData();
form.append("model", "groq/whisper-large-v3-turbo");
form.append("file", new Blob([await readFile("audio.mp3")]), "audio.mp3");
const r = await fetch(`${process.env.KEIROUTER_URL}/v1/audio/transcriptions`, {
method: "POST",
headers: { "Authorization": `Bearer ${process.env.KEIROUTER_KEY}` },
body: form,
});
const { text } = await r.json();
console.log(text);
Response shape
Default (response_format=json):
{ "text": "Hello, this is the transcription.", "language": "en", "duration": 3.2 }
response_format=text returns the transcript as text/plain. Segment timestamps (verbose_json) and subtitle formats (srt, vtt) are not passed through; any other format comes back inside the text field.
Provider quick reference
| Provider | model |
Notes |
|---|---|---|
| OpenAI | whisper-1, gpt-4o-transcribe, gpt-4o-mini-transcribe |
Native OpenAI shape |
| Groq | whisper-large-v3, whisper-large-v3-turbo |
Fastest; OpenAI shape |
| Custom provider | <provider>/<model> |
Any server that speaks OpenAI /audio/transcriptions |
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 Changed · -2 lines · +1 tokens per session 3dbe3b4dc738
- 1mo ago First seen · 78 lines · 57 tokens per session scan A 6831e22e41d1
keirouter-stt is a skill published in the GitHub repository mydisha/keirouter (154 stars, last pushed 5d ago), licensed MIT. It adds 58 tokens to every session and 783 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
caveman-setup
Wire a repository through the Caveman Cloud gateway so every LLM request is measured, with no behavior change. Use for "set up caveman" or adding LLM spend observability.
caveman-discover
Find and label every LLM workflow in the repository so Caveman Cloud groups spend by workflow instead of one bucket. Use for "discover workflows" or breaking LLM spend down by workflow.
omni-inference
The core OpenAI-compatible inference endpoints: chat completions, embeddings, images, audio (TTS/STT), moderations, rerank, and the Responses API. The primary integration surface for AI agents.
cavecrew
When to delegate to cavecrew-investigator (locate code), cavecrew-builder (1-2 file edit) or cavecrew-reviewer (diff review) instead of working inline or using Explore. Their output is compressed, so main context lasts longer.
caveman-evidence-review
Read-only review of Caveman Cloud evidence: cost, Cave Score, workflows, traces, latency, errors, routing, savings. Use when asked what Caveman found or where LLM spend goes.
caveman-manage
Inspect Caveman Cloud's experiment lifecycle and block unsafe execution. Use when asked to start, approve, cancel, promote or roll back a Caveman experiment.