Content Pruning Review
Review and prune website content. Use for removing generic copy, duplicate sections, weak feature cards, unsupported claims, placeholder content, bloated FAQs, repeated testimonials, excessive prose, stale pages, low-value sections, and AI-generated filler before final visual polish.
Install
$ agentstack add skill-thegoat395-codex-skills-content-pruning-review ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Content Pruning Review
Use this skill to make the page sharper by deleting what dilutes the main decision path.
Workflow
- Inspect the page goal, audience, offer/product reality, source copy, proof, IA, forms, SEO/social metadata, brand voice, and verification commands.
- Read [Content Pruning Review Guide](references/content-pruning-review-guide.md) before changing headings, sections, CTAs, metadata, labels, or UI copy.
- Draft from evidence: real product/service facts, proof, objections, user tasks, and constraints.
- Replace generic claims with specific language or remove them.
- Inspect the rendered page on desktop and mobile for hierarchy, wrapping, repetition, and whether copy still matches the UI.
Always Protect
- Inspect before writing or coding: audience, offer, product/service reality, source copy, assets, proof, claims, conversion goal, information architecture, SEO/metadata needs, forms, legal/trust constraints, and available verification commands.
- Use copy to make the design more specific: every headline, label, CTA, section, and proof point should reduce ambiguity rather than decorate the layout.
- Avoid generic premium language, vague transformation claims, placeholder testimonials, fake metrics, empty adjectives, and hero copy that could fit any brand.
- Write in the user's domain vocabulary while keeping language clear, concise, concrete, and scannable.
- Tie claims to proof: screenshots, artifacts, reviews, credentials, specifications, case studies, pricing terms, process details, or real outcomes.
- Keep navigation, forms, pricing, FAQs, and CTAs honest and task-focused; do not create dark patterns or hide material terms.
- Respect accessibility and structure: semantic headings, meaningful link/button text, labeled forms, useful error/help text, readable line lengths, and no critical copy trapped in images.
- Before delivery, inspect rendered desktop and mobile copy for hierarchy, wrapping, overflow, repetition, and whether the words still match the implemented page.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: TheGoat395
- Source: TheGoat395/Codex-Skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.