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

77%

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

Fix and improve this skill with Tessl

tessl review fix ./18-multimodal/audiocraft/SKILL.md

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 executable, copy-paste-ready examples and a reasonable reference split, but it is verbose from repeated near-identical code blocks, lacks validation checkpoints in its batch workflow, and keeps a lot of peripheral content inline rather than in the provided reference files.

Suggestions

Consolidate the repeated MusicGen/AudioGen generate-and-save blocks into one parameterized example and refer back to it, removing the redundancy.

Add explicit validation/verification steps to the batch sound-design workflow (e.g. confirm output files exist and are non-empty, handle generation errors per-spec with retry) to lift workflow clarity.

Move the EnCodec, MusicGen-Style, performance-optimization, and common-issues sections into the existing reference files to keep SKILL.md a lean overview.

Dimension	Reasoning	Score
Conciseness	The body is code-heavy and actionable rather than padded with prose Claude already knows, but ~8 near-identical MusicGen/AudioGen load-configure-generate-save blocks repeat the same pattern, so it could be tightened considerably; this matches 'mostly efficient but could be tightened' rather than 'every token earns its place'.	2 / 3
Actionability	It provides fully executable, copy-paste-ready code with real imports, model IDs, generation parameters, and save calls throughout, matching the 'fully executable code/commands; copy-paste ready' anchor.	3 / 3
Workflow Clarity	Workflows are sequenced (load → configure → generate → save) but the batch sound-design workflow performs repeated generation with no validation or verification checkpoints, and the rubric caps batch/destructive operations at 2 when feedback loops are missing.	2 / 3
Progressive Disclosure	Two real one-level-deep references (advanced-usage.md, troubleshooting.md) are clearly signaled in a References section, but the 555-line body keeps substantial content inline (EnCodec, MusicGen-Style, performance optimization, common issues, and overlapping per-model examples) that could be split out, matching 'some structure but content that should be separate is inline'.	2 / 3
	Total	9 / 12 Passed

Description

90%

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

A strong, specific description that clearly answers both what the skill does and when to use it, with good natural trigger terms and a distinct niche. The only deduction is for second-person voice ('Use when you need to'), which the rubric penalizes on specificity.

Suggestions

Rephrase the trigger clause in third person to avoid the voice penalty, e.g. 'Use when the user needs to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation.'

Consider adding common phrasing variants such as 'text-to-audio' or 'generate sound effects' to broaden natural trigger coverage.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions ('generate music from text descriptions, create sound effects, or perform melody-conditioned music generation'), which is base-3 material, but the second-person phrasing 'Use when you need to' triggers the rubric's voice penalty (−1) for not using third person.	2 / 3
Completeness	It explicitly states what the skill does ('PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen)') and provides an explicit 'Use when...' trigger clause, satisfying both the what and the when.	3 / 3
Trigger Term Quality	It covers natural user-facing terms ('generate music from text descriptions', 'create sound effects', 'melody-conditioned music generation') alongside the product names MusicGen and AudioGen, matching the 'good coverage of natural terms' anchor.	3 / 3
Distinctiveness Conflict Risk	The AudioCraft/MusicGen/AudioGen niche is specific with distinct triggers unlikely to fire for unrelated skills, matching the 'clear niche with distinct triggers' anchor.	3 / 3
	Total	11 / 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 (565 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: Orchestra-Research/AI-Research-SKILLs
Path: 18-multimodal/audiocraft/SKILL.md
Commit: 773a529

Reviewed: about 11 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.