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 sujitnoronha/voicecheck --skill write-voicecheck-testgit clone --depth 1 https://github.com/sujitnoronha/voicecheckWrote 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/sujitnoronha/voicecheck/write-voicecheck-test)<a href="https://agentmods.dev/skills/sujitnoronha/voicecheck/write-voicecheck-test"><img src="https://agentmods.dev/badge/skills/sujitnoronha/voicecheck/write-voicecheck-test.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 5.1 | $0.00085 | $0.01255 |
| Opus 5 | $0.00043 | $0.00628 |
| Sonnet 5 | $0.00017 | $0.00251 |
| Haiku 4.5 | $0.00009 | $0.00126 |
Grade A, and why
write-voicecheck-test 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 8d 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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write a VoiceCheck test
Turn "here's what my agent should do" into a runnable VoiceCheck scenario and, if
they want it in CI, a pytest test. Assume VoiceCheck is already installed; if not,
run /setup-voicecheck first.
A VoiceCheck test is a YAML scenario: a transport (how to reach the agent),
audio providers, and a list of turns, where each turn sends user: text and
lists expect: evaluators that must pass. Run it with voicecheck run, or attach
it to pytest with @pytest.mark.voicecheck("file.yaml").
Shared references (in the sibling setup skill — present when the repo is checked out):
../setup-voicecheck/references/transports.md—transport:block per provider../setup-voicecheck/references/evaluators-and-providers.md— the 13 evaluators + audio providers
The authoritative field list is always voicecheck schema (emits the JSON schema).
Step 1 — Understand what to test
Ask the user (or infer from the agent's code/prompt) before writing YAML:
- Transport: which stack hosts the agent (LiveKit / Daily / VAPI / Retell), or
use
echofor a plumbing-only test. - The conversation: what does a real user say, and what must the agent do each turn? Get 2-4 concrete turns.
- What would count as a failure? This drives evaluator choice — latency budget, required info, forbidden disclosures, tone, tool calls, staying in character.
Step 2 — Choose evaluators per turn
Map each "must / must-not" to an evaluator (full catalog in the reference). Common picks:
| Concern | Evaluator |
|---|---|
| Responds fast enough | latency (max_first_byte_ms) |
| Says / avoids specific words | keyword |
| Did the right thing (free-form) | llm_judge (criteria, min_score) |
| Right tone | emotional_tone (expected_emotions) |
| No system-prompt / tool-name leaks | info_leakage |
| Called a tool (correctly / in order) | tool_called, tool_sequence |
| Stayed in character (roleplay agents) | character_break |
| Quality rubric (PII, policy, brand…) | rubric_judge |
What ships with it
1 file 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.
- 8d ago First seen · 115 lines · 85 tokens per session scan A c4694b23629f
write-voicecheck-test is a skill published in the GitHub repository sujitnoronha/voicecheck (5 stars, last pushed 2mo ago), licensed MIT. It adds 85 tokens to every session and 1,255 once invoked, about $0.0004 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
BrowserStack Cloud Testing
Cloud-based cross-browser and cross-device testing with BrowserStack including Automate, App Automate, Percy visual testing, Observability, and integration with Playwright, Selenium, and CI/CD pipelines.
e2e-alertmanager-test
Render end-to-end previews of what an alert notification will actually look like (plain-text email, HTML email, Slack attachment JSON, raw webhook JSON) by replaying a Prometheus unit-test file's expected alerts through a live Alertmanager. Effectively a "print preview" for alerts. Use to review notification…
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
langsmith-observability
LLM observability platform for tracing, evaluation, and monitoring. Use when debugging LLM applications, evaluating model outputs against datasets, monitoring production systems, or building systematic testing pipelines for AI applications.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
test-warp-ui
Guides testing Warp UI features and changes using the computer use tool. Use this skill only when computer-use testing was requested (explicit request or accepted offer) and the computeruse tool is available to the agent. Covers launching Warp and verifying UI behavior.