youtube-tools
Fetch transcripts and create structured watching guides / summaries for YouTube videos. Use when the user asks to (1) get, fetch, extract, or download a transcript or captions from a YouTube video URL, or (2) summarize, create a watching guide, or produce a structured summary of a YouTube video. Trigger on youtube.com/watch, youtu.be, or youtube.com/shorts links.
95
93%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
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 an excellent skill description that clearly defines concrete capabilities (transcript fetching, structured summaries), provides comprehensive trigger terms covering natural user language variations, and explicitly states both what the skill does and when to use it. The inclusion of specific URL patterns (youtube.com/watch, youtu.be, youtube.com/shorts) further strengthens distinctiveness and trigger accuracy.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Fetch transcripts', 'create structured watching guides', 'summaries for YouTube videos'. Also specifies sub-actions like 'get, fetch, extract, or download a transcript or captions'. | 3 / 3 |
Completeness | Clearly answers both 'what' (fetch transcripts, create structured watching guides/summaries) and 'when' (explicit 'Use when...' clause with numbered trigger scenarios and URL pattern triggers). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: 'transcript', 'captions', 'summarize', 'watching guide', 'YouTube video URL', plus specific URL patterns like 'youtube.com/watch', 'youtu.be', 'youtube.com/shorts'. These are highly natural keywords. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — specifically scoped to YouTube videos with explicit URL patterns. Unlikely to conflict with general summarization or document extraction skills due to the YouTube-specific triggers. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted skill that efficiently teaches Claude how to fetch YouTube transcripts and produce structured watching guides. Its strengths are the concrete CLI commands, the detailed output template with timestamp conversion examples, and lean writing that respects token budget. The main weakness is the lack of error handling guidance in the workflow—particularly for videos without captions or failed fetches.
Suggestions
Add a validation/error-handling step after the transcript fetch (e.g., 'If the command returns no transcript, inform the user that the video likely has no captions and cannot be summarized').
Consider adding a brief note on handling edge cases like live streams, age-restricted videos, or very long videos (>2 hours) that might affect transcript quality or segment count.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is lean and efficient. It doesn't explain what YouTube is, what transcripts are, or how npm works. Every section provides specific, actionable information that Claude wouldn't inherently know (the defuddle CLI, the exact output format, timestamp conversion formula). | 3 / 3 |
Actionability | Provides exact CLI commands (`npx defuddle parse <url> --markdown`), a concrete timestamp conversion formula with a worked example, and a complete copy-paste-ready output template in Markdown. The guidelines for segment count, quote selection, and summary length are specific and actionable. | 3 / 3 |
Workflow Clarity | The watching guide workflow has clear numbered steps (fetch → analyze → produce → save), but lacks validation checkpoints. There's no guidance on what to do if the transcript fetch fails (e.g., no captions available), no verification step after producing the guide, and no error recovery loop for common failure modes like missing captions or incorrect timestamps. | 2 / 3 |
Progressive Disclosure | For a skill of this scope (single tool, two related tasks), the content is well-organized into logical sections (Fetching Transcripts, Creating a Watching Guide) with clear subsections. No bundle files are needed, and the content doesn't warrant splitting into multiple files. The structure supports easy scanning and discovery. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- ceshine/ceshine-agent-skills
- Commit
348c2c1
- 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.