Browse Skills

Finishing a Development Branch verifies tests, detects the git environment, and presents merge, pull request, or keep-as-is options to complete development work. It cleans up worktrees and branches after local merges or confirmed discards.

CLAUDE.md Review audits a CLAUDE.md file for vague directives, unnamed files, stale facts, and context bloat, and proposes concrete rewrites. Use it when asked to review, improve, shrink, or fix a CLAUDE.md, or when Claude Code results feel inconsistent.

CCC Code Search indexes the codebase and provides semantic search over code, so an agent can find relevant code by concept or behavior. Use it when code search is needed, after significant code changes, or when the user asks about ccc or cocoindex-code.

Accessibility audits and improves web accessibility following WCAG 2.2 guidelines. Use when asked to improve accessibility, perform an a11y audit, ensure WCAG compliance, check screen reader support, keyboard navigation, or make a site accessible.

Core Web Vitals optimizes LCP, INP, and CLS to improve page experience and Google Search ranking. Use it when asked to fix LCP, reduce CLS, optimize INP, or resolve page experience and layout shift issues.

SEO optimizes pages for search engine visibility using technical SEO, on-page optimization, and structured data. Use when asked to improve SEO, fix meta tags, add structured data, or optimize sitemaps.

Web Best Practices applies modern web development best practices for security, compatibility, and code quality. Use it when asked to apply best practices, audit security, modernize code, review code quality, or check for vulnerabilities.

Web Performance improves site speed by optimizing loading, runtime efficiency, and resource delivery. Use it when asked to speed up a site, reduce load time, improve page speed, or run a performance audit.

Web Quality Audit runs a comprehensive Lighthouse-based audit covering performance, accessibility, SEO, and best practices. Use it when someone asks to audit a site, review web quality, run a Lighthouse audit, check page quality, or optimize a website.

Notion-as-source-of-truth dispatch board for managing AI agency workflows. Capture chat to Notion, manage tasks through Suggestion-to-Done pipeline, and batch execute with per-task model selection.

Anti-Deception calls a scaffold when a request pressures the assistant to agree, soften an honest assessment, or validate something without evidence. Use it to resist sycophancy, fabricated agreement, and other honesty failures in high-pressure prompts.

Performs multi-level code review (L1-L5) on recent changes for quality, specification compliance, security, and performance. When issues are found, it offers to automatically apply fixes.