dotty-av-test

dotty-av-test is a skill for Claude Code, Codex from BrettKinny/dotty-stackchan. It costs 64 tokens per session (807 once invoked), scanned B, original, MIT.

A local test procedure for a physical Dotty robot that plays spoken prompts through speakers while a camera records its video and room audio. It uses the repository's existing script and connected audio and video hardware.

In plain words
What is it for?
Use it to test wake-word detection, speech recognition, spoken replies, facial or other expressions, and complete voice interactions on the test bench.
Why use it?
It provides a repeatable way to check the robot's real-world voice behavior instead of testing only software components. Recording is kept local unless sharing is separately requested.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Codex.

Good fit Use it to test wake-word detection, speech recognition, spoken replies, facial or other expressions, and complete voice interactions on the test bench.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/brettkinny/dotty-stackchan/dotty-av-test
Install

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.

Any agent
npx skills add BrettKinny/dotty-stackchan --skill dotty-av-test
Clone the repo
git clone --depth 1 https://github.com/BrettKinny/dotty-stackchan

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for dotty-av-test

README.md
[![agentmods](https://agentmods.dev/badge/skills/brettkinny/dotty-stackchan/dotty-av-test/github.svg)](https://agentmods.dev/skills/brettkinny/dotty-stackchan/dotty-av-test)
Your own site
<a href="https://agentmods.dev/skills/brettkinny/dotty-stackchan/dotty-av-test"><img src="https://agentmods.dev/badge/skills/brettkinny/dotty-stackchan/dotty-av-test/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.

agentmods 80×15 button for dotty-av-test

Your own site · 80×15
<a href="https://agentmods.dev/skills/brettkinny/dotty-stackchan/dotty-av-test"><img src="https://agentmods.dev/badge/skills/brettkinny/dotty-stackchan/dotty-av-test.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 64 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 807 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00064 $0.00807
Opus 5 $0.00032 $0.00404
Sonnet 5 $0.00013 $0.00161
Haiku 4.5 $0.00006 $0.00081

Measured yesterday against content hash 224ee4b9975f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-19, from the pricing page.

Security

Grade B, and why

dotty-av-test 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 yesterday.

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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

installed with `sudo pacman -S --needed espeak-ng`; the human may need to run
skills/dotty-av-test/SKILL.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Dotty A/V Test

Use the repository harness at scripts/dotty-av-test.sh; do not recreate its FFmpeg, ALSA, PipeWire, volume-safety, or verification logic ad hoc. Read docs/dotty-av-tests.md when choosing prompts or running a multi-case session.

Before playback

  1. The user must have explicitly requested recording. Keep captures local unless they separately request sharing or upload.
  2. Run scripts/dotty-av-test.sh devices. Prefer the stable C920 index-0 path and its hw:C920,0 microphone; index 1 is commonly metadata-only.
  3. Run scripts/dotty-av-test.sh volume and let the user choose or confirm the speaker level, then run speaker-test. Do not infer that a successful playback process means the user heard a comfortable level.
  4. Confirm espeak-ng, FFmpeg, pw-play, pactl, v4l2-ctl, and ALSA are available. On this Omarchy workstation, the missing TTS dependency is installed with sudo pacman -S --needed espeak-ng; the human may need to run that command because sudo requires their password.

The currently observed hardware is a Logitech C920 (1280x720 MJPEG video; 32 kHz, two-channel S16_LE microphone) and a Volt 4 default PipeWire sink. Rediscover rather than assuming those devices remain connected.

Run and evaluate

Run one test at a time:

scripts/dotty-av-test.sh run "Hi E S P. What is your name?" 20

The harness creates a timestamped MP4 under uat-sessions/<date>/av/, provides two seconds of pre-roll, plays offline TTS, leaves the configured response window, then reports stream metadata and audio levels.

Inspect representative frames when visible state or expression matters. Treat these as separate assertions:

  • capture passed: non-zero video and audio streams with expected durations;
  • playback passed: the prompt was audibly clear at the intended position;
  • interaction passed: Dotty entered listening, understood the prompt, replied, animated appropriately, and recovered to idle.

Never report the interaction as passed merely because the MP4 and audio signal exist. Record silence/non-response as an interaction failure or deferred diagnostic while preserving a successful capture result.

Read the full file on GitHub · 71 lines

Changes

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.

  1. yesterday First seen · 71 lines · 64 tokens per session scan B 224ee4b9975f

Subscribe to this mod's changes

dotty-av-test is a skill published in the GitHub repository BrettKinny/dotty-stackchan (111 stars, last pushed 2d ago), licensed MIT. It adds 64 tokens to every session and 807 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-18.