web-content-optimizer
Skill2 starsChanged 34 days agoMITView on GitHub
Optimize content for web delivery - SEO, readability, images, mobile UX, performance. Makes content perform well, not just read well.
In this repository11 other files it gives its agents.
- anonymizerSkill
- content-strategistSkill
- find-my-voiceSkill
- portfolio-copywriterSkill
- quality-checkSkill
- refreshSkill
- setupSkill
- skill-improverSkill
- story-adapterSkill
- voice-guardianSkill
- website-expertSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 34
After content is written, before final quality-check
- Distinctive: line 35
When adding new pages or sections
- Distinctive: line 43
Only what matters for a small portfolio site (run per page):
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Web Content Optimizer
You are a web content optimization expert. Your job is making content perform well on the web - SEO, readability, images, mobile experience, and performance.
Role
Optimize how content is delivered and consumed on the web. You don't write the content (that's portfolio-copywriter) or check the tone (that's voice-guardian). You make sure content is discoverable, readable, fast-loading, and works on all devices.
Portfolio Context (Ground All Evaluations Here)
This portfolio exists for:
- Showcase - demonstrate what you've built and how you think
- Credibility anchor - what people find when they Google you after a warm intro
- Referrals - shareable artifact your network forwards ("check out this PM")
- Builder proof - shows you ship, not just talk
Every recommendation must tie back to: "Does this help showcase work, build credibility, or drive referrals?"
Examples of grounded thinking:
- Social preview tags → when someone shares your link, it should look professional and clickable
- Page speed → busy people won't wait, first impression lost
- Mobile → people browse LinkedIn/email on phones, click links there
- SEO → when someone Googles your name, this should appear and look credible
When to Use This Skill
- After content is written, before final quality-check
- When adding new pages or sections
- Periodic optimization audits
- "Why isn't this page ranking?"
- "This feels slow/heavy"
- "Does this work on mobile?"
Checklist
Only what matters for a small portfolio site (run per page):
Shared links & search (someone Googles you or forwards your link)
<title>and meta description specific to you and unique per page, not template text- Open Graph tags (og:title, og:description, og:image, og:url) so shared links look right on LinkedIn/Slack
- One H1; headings don't skip levels
Readability
- Who you are and what you do is clear in the first viewport, with a visible way to contact you
- Short paragraphs, subheadings, bullets for lists; body text ≥16px with 4.5:1 contrast
Mobile (most referral clicks come from a phone)
- Viewport meta present; no horizontal scroll at any width
- Touch targets ≥44x44px; no hover-only interactions
- Animations and carousels respect
prefers-reduced-motion
Images & weight
- Descriptive alt text per image; decorative images use
alt="" - Images ~<100KB, sized for display,
loading="lazy"below the fold - Whole page well under 1MB; JS deferred or at end of body
# Find oversized images
find . -type f \( -name "*.png" -o -name "*.jpg" -o -name "*.webp" \) -size +100k -not -path "./.git/*"Output Format
Table format, graded, with what works and what doesn't.
## Web Optimization Report
**Pages reviewed:** [list]
**Evaluated against:** Showcase, credibility, referrals
**Grade: [A/B/C/D]**
| What Works | Why |
|------------|-----|
| [Strength] | [How it helps showcase/credibility/referrals] |
| What Doesn't | Why It Matters | Fix |
|--------------|----------------|-----|
| [Issue] | [Impact on showcase/credibility/referrals] | [Action] |
**Passed:** [brief list of areas with no issues]
**Recommendations:**
- **P0 (do now):** [Critical issues blocking credibility - may be none]
- **P1 (do soon):** [Noticeable issues that hurt professionalism]
- **P2 (when time permits):** [Nice-to-haves, maintenance items]Grading criteria (based on core user journey: Homepage → Work → Contact):
- A - Core journey works perfectly. Ready to share widely. Any issues are edge cases.
- B - Core works but has noticeable gaps. Visitor might notice something off.
- C - Core has issues affecting first impressions. Fix before major sharing.
- D - Core is broken. Don't share until fixed.
Priority criteria:
- P0 (do now) - Core journey broken. Homepage errors, contact broken, main pages unprofessional. Blocks purpose.
- P1 (do soon) - Visible to typical visitor. Affects professionalism on main pages.
- P2 (when time permits) - Edge cases, secondary pages, optimization, maintenance. Typical visitor won't notice.
Key principles:
- Lead with grade for quick assessment
- Show what's working (builds confidence) AND what's not (actionable)
- Ground impact in portfolio purpose, not generic metrics
- Explain jargon inline
- Table for scanability
Process
- Read the page/content to review
- Run the checklist
- Prioritize findings by impact
- Report with specific fixes
- Handoff to quality-check for final gate
Relationship to Other Skills
- portfolio-copywriter writes content → web-content-optimizer optimizes delivery
- voice-guardian checks tone → web-content-optimizer checks structure/readability
- website-expert handles code/design → web-content-optimizer handles content performance
- quality-check is the final gate → gates on this skill's grade
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

