Sync skills from anthropics/skills: claude-academy-guide (new), frontend-design (updated)

[davila7]

Summary

Scheduled sync against https://github.com/anthropics/skills, comparing all 18 upstream skill folders against cli-tool/components/skills/.

• Added ai-research/claude-academy-guide — new upstream skill (Apache 2.0), no local equivalent existed. Recommends Claude Academy courses/tutorials when a user asks how to use a Claude product. Placed under ai-research alongside the existing claude-code-guide (same "product usage guide" shape).
• Updated creative-design/frontend-design/SKILL.md — upstream replaced the content with a substantially revised version (design process, restraint/self-critique, writing-in-design guidance). Local copy was stale; synced to match upstream verbatim.
• Fixed a real bug found by component-reviewer: claude-academy-guide's upstream description: > (YAML folded block scalar) breaks generate_components_json.py's line-based frontmatter parser — it would have extracted the literal description ">" into the catalog (confirmed this already silently affects 8 pre-existing skills, e.g. productivity/devil). Reformatted to a single-line quoted scalar, consistent with how other long descriptions are already written elsewhere in this repo, so both real YAML parsing and the repo's naive extractor produce the correct text.
• Added both skills to cli-tool/components/skills/ANTHROPIC_ATTRIBUTION.md.
• Regenerated the catalog (python scripts/generate_components_json.py, then copied docs/components.json → dashboard/public/components.json) — this also refreshes download-count stats across all component types, which is expected from re-running the generator.

Not changed (checked, no action needed)

• docx, pdf, pptx, xlsx — content drift exists upstream, but ANTHROPIC_ATTRIBUTION.md explicitly documents these as intentional point-in-time snapshots that are "NOT actively maintained," so left untouched.
• claude-api (development/claude-api) — local SKILL.md is a heavily customized rewrite (different trigger heuristics/structure) unrelated to the vanilla upstream content; treated as locally customized and left alone.
• productivity/skill-creator — has local-only files (output-patterns.md, workflows.md, scripts/init_skill.py) indicating a customized fork; left alone.
• pptx/xlsx/docx "office" helper scripts — local-only files (rearrange.py, replace.py, merge_runs.py, etc.) suggest local customization of the shared office library; covered by the snapshot policy above regardless.
• mcp-builder, development/skill-creator, development/web-artifacts-builder — only trivial/reversed diffs (e.g. local already references newer model names than upstream, or a placeholder copyright line); not real updates, skipped.
• All other skills diffed identical to upstream.

Validation

• Both migrated/updated skills reviewed by the component-reviewer agent (frontmatter, naming, security, category placement) — no hardcoded secrets or absolute paths, valid YAML, correct categories confirmed against existing siblings.
• Catalog regeneration verified: claude-academy-guide and frontend-design both resolve to correct category/description in docs/components.json post-fix.

---

Generated by Claude Code

[github-actions]

👋 Thanks for contributing, @davila7!

This PR touches cli-tool/components/** and has been marked review-pending.

What happens next

1. 🤖 Automated security audit runs and posts results on this PR.
2. 👀 Maintainer review — a human reviewer validates the component with the component-reviewer agent (format, naming, security, clarity).
3. ✅ Merge — once approved, your PR is merged to main.
4. 📦 Catalog regeneration — the component catalog is rebuilt automatically.
5. 🚀 Live on aitmpl.com — your component appears on the website after deploy.

While you wait

• Check the Security Audit comment below for any issues to fix.
• Make sure your component follows the contribution guide.

This is an automated message. No action is required from you right now — a maintainer will review soon.

[github-actions]

🎉 SkillSpector Security Scan

Static analysis (--no-llm) by SkillSpector (NVIDIA, Apache-2.0).

Status: ✅ PASSED

Metric	Count
Skills scanned	2
🔴 HIGH/CRITICAL (score > 50)	0
🟡 MEDIUM	0
🟢 LOW / clean	2

No HIGH/CRITICAL findings. ✅

---

Generated by SkillSpector. Score bands: 0-20 LOW, 21-50 MEDIUM, 51-80 HIGH, 81-100 CRITICAL.

[github-actions]

⚠️ Security Audit Report

Status: ❌ FAILED

Metric	Count
Total Components	769
✅ Passed	358
❌ Failed	411
⚠️ Warnings	977

❌ Failed Components (Top 5)

Component	Errors	Warnings	Score
vercel-edge-function	3	4	81/100
prompt-engineer	2	0	90/100
neon-expert	2	2	88/100
agent-overview	2	1	89/100
unused-code-cleaner	2	1	89/100

...and 406 more failed component(s)

---

📊 View Full Report for detailed error messages and all components

[davila7]

The ❌ Security Audit failure here is pre-existing on main, not caused by this PR. I ran npm run security-audit:ci against origin/main directly: 769 total components, 355 passed, 414 failed — essentially the same numbers as this PR's report (358/411), and none of the top failing components listed (vercel-edge-function, prompt-engineer, neon-expert, agent-overview, unused-code-cleaner) are touched by this PR's diff (which only adds/modifies ai-research/claude-academy-guide and creative-design/frontend-design). Will keep an eye out for a base-branch recovery and re-run if that happens; no action needed from this PR otherwise.

---

Generated by Claude Code

[cubic-dev-ai]

This PR has 50,193 reviewable changed lines after ignored/generated files are excluded, above cubic's default 50,000-changed-line automatic review limit.

Most of the diff comes from:

• dashboard/public/components.json (~46,120 changed lines)
• docs/components.json (~3,652 changed lines)
• cli-tool/components/skills/ai-research/claude-academy-guide/LICENSE.txt (~202 changed lines)
• cli-tool/components/skills/ai-research/claude-academy-guide/SKILL.md (~133 changed lines)
• cli-tool/components/skills/creative-design/frontend-design/SKILL.md (~65 changed lines)

Comment @cubic-dev-ai review this to review it anyway. If the largest files are generated or fixture data, add them to your ignored files in review settings or ignorePatterns in cubic.yaml - cubic will then review the rest automatically. You can also raise this limit in review settings.

[greptile-apps]

Greptile Summary

The PR synchronizes Anthropic’s new Claude Academy guide and revised frontend-design skill, then updates attribution and generated catalog assets.

• Adds the claude-academy-guide skill and license under ai-research.
• Updates the frontend-design skill to the latest upstream guidance.
• Corrects quoted-description extraction and regenerates documentation and dashboard catalog artifacts.

Confidence Score: 5/5

The PR appears safe to merge.

No blocking failure remains.

Important Files Changed

Filename	Overview
cli-tool/components/skills/ai-research/claude-academy-guide/SKILL.md	Adds the Claude Academy recommendation skill with valid single-line frontmatter and detailed catalog-use safeguards.
cli-tool/components/skills/creative-design/frontend-design/SKILL.md	Replaces the prior frontend-design guidance with the revised upstream design process and self-critique instructions.
scripts/generate_components_json.py	Adds scalar-delimiter cleanup that resolves the previously reported quotation-mark leakage in generated descriptions.
docs/components.json	Regenerates the documentation catalog with the new skill, updated metadata, and correctly decoded Claude Academy description.
dashboard/public/components.json	Refreshes the dashboard catalog and associated generated artifacts to match the canonical component sources.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart LR
  Skills[cli-tool/components/skills] --> Generator[scripts/generate_components_json.py]
  Generator --> Docs[docs/components.json]
  Generator --> Dashboard[dashboard/public catalog and search artifacts]
  Dashboard --> Users[Dashboard browsing and search]

Loading

_{Reviews (2): Last reviewed commit: "fix: Strip YAML quote delimiters from ca..." | Re-trigger Greptile}