agentsclimarketplace

Audio recap

Skill dmltdev/void-grimoire/skills/audio-recap

Turn changes made by an agent into a concise Markdown script for text-to-speech. Use only when the user explicitly asks for an audio recap, spoken recap, TTS-ready change summary, or invokes /audio-recap.From its SKILL.md

Install
npx -y skills add dmltdev/void-grimoire --skill audio-recap

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.8 KB, as published. Nobody here has run it

Audio Recap

Create a TTS-ready .md file that explains completed agent work out loud.

Use this only when explicitly invoked. Do not auto-run after every change.

The listener may be a non-native English speaker. Optimize for clear spoken understanding, not visual scanning.

Input Contract

Use the current session first:

  • user request
  • decisions made
  • files changed
  • commands run
  • verification observed
  • unfinished work

Use git diff or stat when available to ground changed files. Do not require a commit, branch, or pull request.

If the user gives a PR, branch, commit, or diff, recap that target instead.

Output Location

Write recaps under docs/audio-recaps/<slug>.audio-recap.md. Build <slug> from the work title in kebab-case. If the file exists, suffix with the next integer, like <slug>.audio-recap-2.md. Never overwrite.

Writing Rules

  • Use B2 English.
  • Use short sentences.
  • Use simple words when possible.
  • Explain important jargon once.
  • Avoid idioms, jokes, metaphors, and dense technical phrasing.
  • Prefer spoken prose over visual Markdown.
  • Use headings and short paragraphs.
  • Use tiny checklists only when sequence matters.
  • Do not use tables.
  • Do not use code blocks unless the exact command or prompt must be copied.
  • Mention only key file paths, and speak why they matter.
  • Keep the script about 450-1,000 words, roughly 3-7 minutes spoken.

Truth Rules

  • Include only changes that were actually made or observed.
  • Include verification only when the command or check was observed.
  • If no verification was run, say: "This change was not verified in this session."
  • Do not claim performance, security, parity, or integration success unless it was checked.
  • Separate finished work from remaining work.

Content Shape

# Audio Recap: {Work Title}

## Short version
{2-4 sentences explaining what changed and current status.}

## What the user asked for
{Plain-language request summary.}

## What changed
{Main changes by feature/system area, not line-by-line.}

## Important files
{Only key files, with why they matter.}

## Verification
{Observed checks and results, or explicit not-verified sentence.}

## Risks and follow-up
{Known risks, limitations, or next steps.}

Omit empty sections. Keep the story linear.

Final Handoff

After writing the file, always print:

Audio recap saved to <path>.

To create spoken audio, paste:
Read @<path> and use /using-elevenlabs-tts to create spoken audio from it.

Use the exact generated path.

Keep looking

Skills are one crate of 608,475. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.