voice-refine
Transform verbose voice input into structured, token-efficient Claude prompts. Use when cleaning up voice memos, dictation output, or speech-to-text transcriptions that contain filler words, repetitions, and unstructured thoughts.
82
78%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./examples/skills/voice-refine/SKILL.mdQuality
Discovery
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.
This is a well-crafted skill description that clearly communicates its purpose, includes natural trigger terms across multiple variations of voice/speech input, and explicitly states both what it does and when to use it. The description is concise without being vague, and it carves out a distinct niche that minimizes conflict with other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists specific concrete actions: 'Transform verbose voice input into structured, token-efficient Claude prompts' and describes the types of content it handles (filler words, repetitions, unstructured thoughts). Multiple specific capabilities are named. | 3 / 3 |
Completeness | Clearly answers both 'what' (transform verbose voice input into structured, token-efficient prompts) and 'when' (explicit 'Use when' clause covering voice memos, dictation output, speech-to-text transcriptions with filler words, repetitions, and unstructured thoughts). | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms users would say: 'voice input', 'voice memos', 'dictation output', 'speech-to-text', 'transcriptions', 'filler words', 'repetitions'. Good coverage of variations a user might naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Occupies a clear niche at the intersection of voice/speech-to-text processing and prompt optimization. The specific focus on voice input cleanup for Claude prompts is highly distinctive and unlikely to conflict with general text editing or prompt engineering skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is well-structured and clearly organized with good progressive disclosure, but falls short on actionability by not showing a complete before/after transformation example. The pipeline is clearly sequenced but lacks validation steps, and some content (compression metrics, concept explanations about what to filter) could be tightened. The biggest gap is the missing output for the French voice input example, which would make the skill significantly more actionable.
Suggestions
Add the expected refined output for the French voice input example to show a complete before/after transformation, making the skill copy-paste demonstrable.
Remove or condense the compression metrics table — Claude doesn't need numeric targets to perform text refinement effectively.
Add a validation/feedback step in the pipeline, e.g., after COMPRESS, verify that all technical requirements from the original are preserved in the output.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient but includes some unnecessary elements like the compression metrics table (Claude doesn't need target percentages to perform well) and the flags table which describes CLI features that aren't clearly tied to an implementation. The filtering rules section is useful but could be more compact. | 2 / 3 |
Actionability | The transformation pipeline and output format template are helpful, but the skill lacks executable code or concrete before/after examples showing the actual transformation result. The input example is provided but the corresponding output is missing, making it unclear exactly what the refined prompt should look like. | 2 / 3 |
Workflow Clarity | The 4-step pipeline (DEDUPE → EXTRACT → STRUCTURE → COMPRESS) provides a clear sequence, but there are no validation checkpoints or feedback loops. The --confirm flag implies a review step but the workflow doesn't explicitly describe what happens if the user rejects the refinement or how to iterate. | 2 / 3 |
Progressive Disclosure | Content is well-organized with clear sections, appropriate length for a SKILL.md, and references to external files (guide/ai-ecosystem.md, examples/before-after.md) that are one level deep and clearly signaled. The structure supports easy scanning. | 3 / 3 |
Total | 9 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Commit
746adc8
- Reviewed
Table of Contents
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.