feat: optimize 5 lowest-scoring skills + add Tessl review action#2
feat: optimize 5 lowest-scoring skills + add Tessl review action#2rohan-tessl wants to merge 2 commits into
Conversation
Optimized the 5 lowest-scoring skills (out of 25) using tessl skill review. Restructured content following CONTRIBUTING.md guidelines: SKILL.md under 500 lines with detailed content moved to references/. | Skill | Before | After | Change | |---------------------|--------|-------|--------| | form-cro | 59% | 88% | +29% | | signup-flow-cro | 62% | 89% | +27% | | programmatic-seo | 64% | 83% | +19% | | seo-audit | 64% | 94% | +30% | | marketing-psychology| 67% | 94% | +27% | Key improvements across all skills: - Descriptions rewritten with concrete action verbs and Use-when clauses - Added explicit numbered workflows with validation checkpoints - Moved verbose reference content to references/ directories - Added concrete worked examples and output templates - Trimmed generic knowledge Claude already understands
Adds .github/workflows/skill-review.yml that automatically runs tessl skill review on any PR touching SKILL.md files and posts scores as a PR comment. No Tessl login needed — uses GITHUB_TOKEN. Non-blocking by default (fail-threshold: 0).
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request significantly restructures several marketing and SEO skills by moving detailed tactical guidance into a new references/ directory and streamlining the core SKILL.md files into actionable workflows. Key changes include the introduction of comprehensive checklists for SEO audits, experiment libraries for signup flows, and a structured mapping of marketing psychology models to specific business challenges. The review feedback focuses on refining technical diagnostic commands, clarifying navigation requirements for large-scale programmatic SEO, and ensuring A/B testing durations follow statistical best practices.
| ### E-E-A-T Signals | ||
| **Check page title and meta description:** | ||
| ```bash | ||
| curl -s https://example.com | grep -iE '<title>|<meta name="description"' |
There was a problem hiding this comment.
The grep command for the title and meta description is quite fragile. It will fail if the tags have attributes (like dir or lang on the title) or if the order of attributes in the meta tag is different (e.g., content before name). Consider using a more flexible regex to ensure these common variations are captured during diagnostics.
| curl -s https://example.com | grep -iE '<title>|<meta name="description"' | |
| curl -s https://example.com | grep -iE '<title|<meta.*name="description"' |
| - **Cross-links**: Related spokes link to each other | ||
|
|
||
| ### Post-Launch Monitoring | ||
| Ensure every page is reachable from main navigation, included in XML sitemap, and has breadcrumbs with structured data. |
There was a problem hiding this comment.
The instruction "Ensure every page is reachable from main navigation" is ambiguous for programmatic SEO projects that often involve thousands of pages. It is physically impossible to include all pages in the main navigation. It would be clearer to specify that the hub pages should be in the main navigation, and all spoke pages should be reachable through the site's internal linking architecture.
| Ensure every page is reachable from main navigation, included in XML sitemap, and has breadcrumbs with structured data. | |
| Ensure every hub page is reachable from main navigation, all spoke pages are connected via internal linking, and all are included in the XML sitemap with breadcrumbs and structured data. |
| - Prioritize high-volume patterns for initial indexation | ||
| - Noindex very thin or low-demand variations | ||
| - Use separate sitemaps per page type | ||
| - Manage crawl budget — submit pages in batches |
There was a problem hiding this comment.
"Submit pages in batches" is a bit vague in an SEO context. If this refers to the Google Indexing API, it is generally restricted to specific content types. If it refers to publishing, it is a valid strategy for monitoring quality. Clarifying this to "publish in batches" makes the recommendation more actionable.
| - Manage crawl budget — submit pages in batches | |
| - Manage crawl budget — publish pages in batches to monitor indexation and quality |
| - Recommend A/B test: current layout vs. restructured page | ||
| - Primary metric: trial start rate | ||
| - Secondary metric: Pro plan selection rate | ||
| - Test duration: 2 weeks or 1,000 visitors per variant (whichever comes first) |
There was a problem hiding this comment.
In A/B testing, stopping a test as soon as a visitor count is reached ("whichever comes first") can lead to misleading results due to time-based variations, such as day-of-week effects. It is a best practice to define a minimum duration that covers at least one full business cycle (usually 1-2 weeks) AND a required sample size to ensure statistical validity.
| - Test duration: 2 weeks or 1,000 visitors per variant (whichever comes first) | |
| - Test duration: Minimum 2 weeks and at least 1,000 visitors per variant to ensure statistical validity. |
1 participant
Hey @mysticaltech 👋
I ran your skills through
tessl skill reviewat work and found some targeted improvements. Here's the full before/after:This PR covers your 5 lowest-scoring skill(s) in the repo (out of 25 total). The remaining 20 skills are already scoring well — and the included GitHub Action will review them automatically on future PRs.
What changed in form-cro
references/— field-by-field optimization toreferences/field-optimization.md, experiments toreferences/experiments.mdWhat changed in signup-flow-cro
references/— experiments toreferences/experiments.md, patterns toreferences/flow-patterns.mdWhat changed in programmatic-seo
What changed in seo-audit
references/—references/technical-seo-checklist.mdandreferences/on-page-checklist.mdWhat changed in marketing-psychology
references/—references/mental-models.mdandreferences/pricing-psychology.mdAll changes follow your CONTRIBUTING.md guidelines: SKILL.md files stay well under 500 lines, detailed content lives in
references/, names match directories, and descriptions include trigger phrases.Tessl Skill Review GitHub Action ✅
I've also included a GitHub Action (
.github/workflows/skill-review.yml) that automatically reviews anySKILL.mdchanged in future PRs and posts scores as a PR comment.What this gives you:
tessl skill reviewruns on every PR touchingSKILL.mdGITHUB_TOKENis usedfail-threshold: 70later if you want a hard gate)Want automatic AI optimization on every SKILL.md change? 🚀
The action I've added gives you review scores on PRs. We also have a more powerful variant —
tesslio/skill-review-and-optimize— that can:SKILL.mdPR (requires addingTESSL_API_TOKENas a repo secret)/apply-optimizeInterested? Tick the box and I'll raise a follow-up PR:
tesslio/skill-review-and-optimizeaction so every SKILL.md PR gets AI optimization suggestions + the/apply-optimizeflowHonest 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 and ask it to optimize your skill. Ping me — @rohan-tessl — if you hit any snags.
Thanks in advance 🙏