whisper

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/llm-tools/whisper/SKILL.md

Quality

Content

65%Weight 40%Scale 1-5

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

A highly actionable, code-rich skill body that covers Whisper's common use cases well. Its main weaknesses are duplicated/inlined language content that should delegate to the existing reference file, marketing-style metrics, and missing validation in the batch workflow.

Suggestions

Replace the inlined 'Language support' section with a one-line pointer to references/languages.md (e.g., 'See [languages.md](references/languages.md) for the full 99-language list, per-language tips, and WER tables') to fix the broken progressive-disclosure link.

Remove marketing metrics ('72,900+ GitHub stars', 'Trained on 680,000 hours of audio') from the body; keep only what Claude needs to execute the skill.

Add a validation/error-handling step to the batch-processing loop (e.g., wrap model.transcribe in try/except, check result['text'] is non-empty, log failures) so the batch workflow can score above 3 on workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is mostly actionable code without over-explaining concepts Claude knows, but it carries marketing fluff ('72,900+ GitHub stars', 'Trained on 680,000 hours of audio') and a 'Language support' section that duplicates the existing references/languages.md, so it could be tightened.	3 / 5
Actionability	Provides abundant copy-paste-ready, executable code and CLI commands covering transcription, language/task selection, timestamps, temperature fallback, batch processing, real-time, and GPU use — fully covering the common cases.	5 / 5
Workflow Clarity	Sections are loosely sequenced (install → basic → options → advanced), but the batch-processing loop has no validation or error-handling; per the rubric, batch operations without validation cap workflow clarity at 3.	3 / 5
Progressive Disclosure	A bundle file references/languages.md exists, but the body never links to it — instead it inlines a language list and writes 'Full list: 99 languages total' without pointing to the reference, leaving content that belongs in the separate file inlined and the reference unsignaled.	3 / 5
	Total	14 / 20 Passed

Description

83%Weight 40%Scale 1-5

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, concrete description that names the tool, its capabilities, model sizes, and clear use-case triggers. It could be improved by adding explicit user-mention trigger phrasing and natural synonyms/file extensions.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'transcription, translation to English, and language identification' plus 'speech-to-text, podcast transcription, or multilingual audio processing' — giving comprehensive coverage rather than vague abstraction.	5 / 5
Completeness	Clearly states what the model does (99 languages, six sizes, transcription/translation/identification) and gives a 'Use for...' trigger clause, but the 'when' is framed as use-cases rather than explicit user-mention triggers, so it is not a clean 5.	4 / 5
Trigger Term Quality	Includes natural phrases users say ('speech-to-text', 'podcast transcription', 'multilingual audio processing', 'transcription', 'translation'), but misses common synonyms and file extensions like 'transcribe audio', 'subtitles', or '.mp3/.wav'.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (OpenAI's Whisper speech recognition) with distinct audio/transcription triggers and minimal overlap risk with unrelated skills.	5 / 5
	Total	18 / 20 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
metadata_version	'metadata.version' is missing	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/llm-tools/whisper/SKILL.md
Commit: 543b17f

Reviewed: 2 days 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.