Skip to content

Improve/skill review optimization#1

Open
yogesh-tessl wants to merge 2 commits intoshenron0101:mainfrom
yogesh-tessl:improve/skill-review-optimization
Open

Improve/skill review optimization#1
yogesh-tessl wants to merge 2 commits intoshenron0101:mainfrom
yogesh-tessl:improve/skill-review-optimization

Conversation

@yogesh-tessl
Copy link
Copy Markdown

Hey @shenron0101 👋

I ran your skills through tessl skill review at work and found some targeted improvements. Here's the full before/after:


score_card

Note: Two skills (claude2opencode-skill and pdf-study-qa-skill) originally had no YAML frontmatter, which caused them to fail deterministic validation (scoring 10%). I added minimal frontmatter (name + description) and re-scored to get meaningful baseline scores — the "Before" column reflects those corrected baselines. The claude2opencode-skill name was also changed from claude2opencode to opencode-migrator because claude is a reserved word in skill names.

Changes made

opencode-migrator (claude2opencode-skill) — 51% → 90%

  • Added YAML frontmatter with proper name and description fields
  • Renamed skill from claude2opencode to opencode-migrator to avoid reserved word violation
  • Expanded description with "Use when..." clause and natural trigger terms (migrate, port, switch, .claude, configuration migration)
  • Replaced abstract usage section with a concrete 5-step workflow including validation and verification commands
  • Removed redundant "Description" section (duplicated title) and "Requirements" section (known concepts)
  • Added executable bash examples for config analysis, migration, output validation, and manifest inspection

graph-visualizer — 90% → 90%

  • Converted frontmatter description from YAML folded block scalar (>) to quoted string format
  • Consolidated repeated visibility warnings into a single "Visibility Rules" reference section
  • Consolidated per-variant aesthetic specifications into a single reference table in Step 2
  • Added 3 validation checkpoints (starfield rendering, node visibility, bloom activation)
  • Removed redundant "see per-variant specifications" back-references from later steps

pdf-study-qa — 33% → 86%

  • Added YAML frontmatter with proper name and description fields
  • Expanded description with "Use when..." clause and natural trigger terms (PDFs, .pdf, read PDF, summarize, chat with document)
  • Replaced abstract 4-step workflow with executable Python code (pdfplumber extraction, chunking, Q&A with citations)
  • Added validation checklist (zero-page check, OCR fallback, page count logging)
  • Removed padded sections (Supported PDF Types, generic Features, Requirements)
  • Added error handling reference table covering common failure modes

I also added a lightweight GitHub Action that auto-reviews any skill.md changed in a PR (includes min permissions, uses a pinned action version, only posts a review comment).

this means that it gives you and your contributors an instant quality signal before you have to review yourself (no signup, no tokens needed).

These were straightforward adjustments to align the skill with Anthropic's best practices for what performs well. Full transparency — I work at tessl.io, where we build tooling around this. Not trying to pitch anything, just sharing fixes that were easy to implement. Happy to answer any questions about the changes!

Thanks in advance 🙏

Hey @shenron0101 👋

I ran your skills through `tessl skill review` at work and found some targeted improvements. Here's the full before/after:

| Skill | Before | After | Change |
|-------|--------|-------|--------|
| opencode-migrator | 51% | 90% | +39% |
| graph-visualizer | 90% | 90% | — |
| pdf-study-qa | 33% | 86% | +53% |

> **Note:** Two skills (`claude2opencode-skill` and `pdf-study-qa-skill`) originally had no YAML frontmatter, which caused them to fail deterministic validation (scoring 10%). I added minimal frontmatter (name + description) and re-scored to get meaningful baseline scores — the "Before" column reflects those corrected baselines. The `claude2opencode-skill` name was also changed from `claude2opencode` to `opencode-migrator` because `claude` is a reserved word in skill names.

<details>
<summary>Changes made</summary>

### opencode-migrator (claude2opencode-skill) — 51% → 90%
- Added YAML frontmatter with proper `name` and `description` fields
- Renamed skill from `claude2opencode` to `opencode-migrator` to avoid reserved word violation
- Expanded description with "Use when..." clause and natural trigger terms (migrate, port, switch, .claude, configuration migration)
- Replaced abstract usage section with a concrete 5-step workflow including validation and verification commands
- Removed redundant "Description" section (duplicated title) and "Requirements" section (known concepts)
- Added executable bash examples for config analysis, migration, output validation, and manifest inspection

### graph-visualizer — 90% → 90%
- Converted frontmatter description from YAML folded block scalar (`>`) to quoted string format
- Consolidated repeated visibility warnings into a single "Visibility Rules" reference section
- Consolidated per-variant aesthetic specifications into a single reference table in Step 2
- Added 3 validation checkpoints (starfield rendering, node visibility, bloom activation)
- Removed redundant "see per-variant specifications" back-references from later steps

### pdf-study-qa — 33% → 86%
- Added YAML frontmatter with proper `name` and `description` fields
- Expanded description with "Use when..." clause and natural trigger terms (PDFs, .pdf, read PDF, summarize, chat with document)
- Replaced abstract 4-step workflow with executable Python code (pdfplumber extraction, chunking, Q&A with citations)
- Added validation checklist (zero-page check, OCR fallback, page count logging)
- Removed padded sections (Supported PDF Types, generic Features, Requirements)
- Added error handling reference table covering common failure modes

</details>

Honest disclosure — I work at @tesslio where we build tooling around skills like these. Not a pitch - just saw room for improvement and wanted to contribute.

Want to self-improve your skills? Just point your agent (Claude Code, Codex, etc.) at [this Tessl guide](https://docs.tessl.io/evaluate/optimize-a-skill-using-best-practices) and ask it to optimize your skill. Ping me - [@yogesh-tessl](https://github.com/yogesh-tessl) - if you hit any snags.

Thanks in advance 🙏
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants