audiocraft-audio-generation

AudioCraft: MusicGen text-to-music, AudioGen text-to-sound.

Quality

61%

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 ./skills/mlops/models/audiocraft/SKILL.md

Quality

Content

72%

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 clean progressive disclosure into real reference files, but it is somewhat verbose (redundant usage blocks and over-explained concepts) and its workflows lack explicit validation checkpoints. Tightening repeated sections and adding verify steps to batch operations would lift the weaker dimensions.

Suggestions

Consolidate the duplicated MusicGen basic-usage examples (Quick start vs. MusicGen usage) into a single authoritative block to reduce redundancy.

Remove or condense concepts Claude already knows (e.g., the architecture ASCII diagram and inline definitions of top_k/temperature) to improve token efficiency.

Add explicit validation/feedback checkpoints to the batch-generation and Gradio workflows (e.g., verify output shape and sample rate before saving, retry on failure).

Dimension	Reasoning	Score
Conciseness	Mostly efficient executable content, but it re-explains fundamentals Claude already knows (the T5/decoder architecture diagram, in-comment definitions of top_k/temperature) and duplicates MusicGen basic usage across the Quick start and MusicGen usage sections.	2 / 3
Actionability	Provides fully executable, copy-paste-ready code throughout for MusicGen, AudioGen, EnCodec, batch processing, and a Gradio demo, with specific parameters and model identifiers.	3 / 3
Workflow Clarity	Multi-step workflows (batch sound generation, Gradio demo, music pipeline) are sequenced, but none include explicit validation or validate-fix-retry checkpoints; for batch operations the guidelines cap workflow clarity at 2.	2 / 3
Progressive Disclosure	The SKILL.md body is an overview with two well-signaled, one-level-deep references ([advanced-usage.md], [troubleshooting.md]) that exist as real files, and detailed material is appropriately split out.	3 / 3
	Total	10 / 12 Passed

Description

50%

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 and names concrete capabilities but omits any explicit use-when trigger guidance and covers only two of the toolkit's operations. Adding a "Use when..." clause with broader trigger terms would raise both completeness and trigger quality.

Suggestions

Add an explicit "Use when..." clause naming the situations that should activate this skill (e.g., generating music, sound effects, or audio from text).

Broaden trigger terms to include natural phrasings like "music generation", "sound effects", and "AudioGen" alongside MusicGen.

Mention additional concrete capabilities (melody conditioning, stereo output, EnCodec compression) to improve specificity and distinctiveness.

Dimension	Reasoning	Score
Specificity	Names the AudioCraft domain and two concrete actions ("text-to-music", "text-to-sound") but covers only two operations, not a comprehensive list (melody conditioning, stereo, EnCodec all omitted).	2 / 3
Completeness	States what the skill does but lacks any "Use when..." clause, so the "when to use" half is missing rather than explicit; per guidelines this caps completeness at 2.	2 / 3
Trigger Term Quality	Includes natural terms a user might say ("MusicGen", "text-to-music", "text-to-sound") but misses common variations like "music generation", "sound effects", or "AudioGen" as standalone triggers.	2 / 3
Distinctiveness Conflict Risk	The AudioCraft/MusicGen naming gives a recognizable niche, but the terse description provides no explicit distinguishing triggers and could overlap with broader audio-generation skills.	2 / 3
	Total	8 / 12 Passed

Validation

75%

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 — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (569 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	12 / 16 Passed

Repository: NousResearch/hermes-agent
Path: skills/mlops/models/audiocraft/SKILL.md
Commit: d0d116b

Reviewed: about 13 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.