audiocraft-audio-generation

PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with complete executable examples and useful reference tables, but it is overly long and repetitive and keeps far too much detail inline rather than pushing it to the bundled reference files. Validation/feedback loops are absent for batch and generation workflows.

Suggestions

Collapse the repeated load/configure/generate/save MusicGen snippets into one canonical example and factor the melody, stereo, and style variants into advanced-usage.md to cut the body substantially.

Add explicit validate/feedback steps to the batch and Gradio workflows (e.g., check output shape/sample rate, catch and retry on OOM) so destructive/batch operations have checkpoints.

Move the architecture overview and generation-parameter tables into references/advanced-usage.md, leaving SKILL.md as a concise quick-start overview with signaled links.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with executable code, but the ~550-line body repeats the same load/configure/generate/save boilerplate across many near-identical snippets (e.g., MusicGen basic, medium, melody, stereo all re-instantiate MusicGen) and includes an ASCII architecture diagram and key-features prose that add length without proportional value.	2 / 3
Actionability	Numerous complete, copy-paste-ready executable Python snippets (MusicGen, AudioGen, EnCodec, melody/stereo/style conditioning, batch processing, Gradio demo) plus concrete model names, sample rates, and parameter tables — fully actionable.	3 / 3
Workflow Clarity	Workflows are presented as runnable scripts but lack explicit validation checkpoints or feedback loops; the batch workflow (Workflow 2) writes many files without verify/retry guidance, and error handling is delegated to a generic "Common issues" table rather than inline check->fix->retry steps.	2 / 3
Progressive Disclosure	Two real reference files (advanced-usage.md, troubleshooting.md) are correctly signaled at the end, but the main body itself is a monolithic wall (~550 lines) of inline API reference and many full code examples that should live one level deeper, so structure is only partially offloaded.	2 / 3
	Total	9 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concise, third-person, and clearly answers both what the skill does and when to use it, with concrete actions and natural trigger terms. It is a strong, well-scoped description.

Dimension	Reasoning	Score
Specificity	Names concrete capabilities: "text-to-music (MusicGen) and text-to-sound (AudioGen)", "generate music from text descriptions", "create sound effects", "melody-conditioned music generation" — multiple specific actions, matching the level-3 anchor.	3 / 3
Completeness	Explicitly states what it does ("PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen)") and when to use it via an explicit "Use when you need to..." clause answering both what and when.	3 / 3
Trigger Term Quality	Covers natural user-language terms ("generate music from text descriptions", "create sound effects", "melody-conditioned music generation") alongside the product names MusicGen/AudioGen, giving good coverage of phrases a user would actually say.	3 / 3
Distinctiveness Conflict Risk	The audio-generation niche with named MusicGen/AudioGen triggers is clearly distinct and unlikely to fire for unrelated skills; it occupies a clear, specific niche.	3 / 3
	Total	12 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (566 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/coding/audiocraft/SKILL.md
Commit: 52845c3

Reviewed: about 12 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.