Changelog
Tracks additions, changes, and fixes across major skill library releases for a web consulting project.
What this file does
Tracks additions, changes, and fixes across major skill library releases for a web consulting project.
When to use it
- Reviewing what changed between versions before upgrading
- Checking which external skill sources were added or updated
- Verifying that a specific skill or fix was included in a release
Assumes this stack
Changelog
All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog.
Unreleased
Added
-
frontend-design: Distinctive UI aesthetics from Anthropic
- Anti-AI-slop patterns for memorable interfaces
- Creative typography pairings (avoid Inter/Roboto/Arial)
- Color palettes with dominant + accent approach
- Motion design for high-impact moments
- Component examples: dark tech hero, glass morphism cards, pricing cards
- Quick reference CSS variables system
-
web-design-guidelines: Interface review for accessibility from Vercel
- WCAG 2.1 AA compliance checking
- Semantic HTML validation patterns
- ARIA patterns for common UI components
- Keyboard navigation requirements
- Focus states and color contrast guidelines
- Quick checklist for pre-launch review
-
document-processing: PDF, DOCX, PPTX, XLSX processing from Anthropic
- PDF: pdfplumber text extraction, pypdf merge/split/rotate, reportlab creation, OCR with pytesseract
- DOCX: pandoc conversion, docx-js creation, OOXML tracked changes workflow
- PPTX: markitdown extraction, PptxGenJS creation, template workflows
- XLSX: pandas analysis, openpyxl formulas, financial model standards
-
update.sh: New update script for easy skill management
--pullflag to pull git changes before updating--sync-onlyflag to only sync external skills--forceflag to stash local changes--dry-runflag to preview changes- Automatic lastSync timestamp in .sync-config.json
Changed
-
composer.json: Added composer scripts for skill management
composer skills:install- Install skills to ~/.claude/skillscomposer skills:update- Sync external skills and reinstallcomposer skills:sync- Only sync external skillscomposer skills:list- List all available skills
-
.sync-config.json: Added external sources for new skills
- web-design-guidelines from Vercel
- Reference entries for Anthropic skills (disabled, using local versions)
- Reference entries for marketing-skills (disabled, using local version with supplements)
-
AGENTS.md: Updated with new skills
- Added frontend-design, web-design-guidelines, document-processing to overview
- Added example prompts for new skills
- Added new session profiles (Creative Design, Accessibility Audit, Document Processing)
- Updated compatibility matrix
-
README.md: Updated documentation
- Added new update script documentation
- Added composer scripts documentation
- Added sync-config.json documentation
- Updated skill counts and categories
- Added Anthropic to external skill sources
-
react-best-practices: React and Next.js performance optimization from Vercel Engineering
- 57 rules across 8 priority categories
- Eliminating waterfalls (async/await patterns, Promise.all, Suspense)
- Bundle size optimization (barrel imports, dynamic imports, defer third-party)
- Server-side performance (React.cache, parallel fetching, after() API)
- Client-side data fetching (SWR, event listeners, passive listeners)
- Re-render optimization (memo, functional setState, useTransition, refs)
- Rendering performance (content-visibility, hydration, Activity component)
- JavaScript performance (batching, caching, Set/Map lookups)
- Advanced patterns (event handler refs, init once, useLatest)
-
readiness-report: AI agent readiness assessment from OpenHands
- 9 technical pillars: Style, Build, Testing, Docs, DevEnv, Debugging, Security, Tasks, Analytics
- 5 maturity levels (L1 Sandbox to L5 Collaborative)
- Automated repository analysis scripts
- Prioritized recommendations for improving agent readiness
-
security-incident-reporting: Security incident documentation templates
- NIST/SANS incident report structure
- DDoS post-mortem templates with timeline documentation
- CVE correlation and MITRE ATT&CK TTP mapping
- Blameless root cause analysis (5-Whys)
- SKILL-TYPO3.md: TYPO3 forensics, Security Team communication, PGP templates
-
firecrawl: LLM-optimized web scraping and research from Mendable AI
- Replaces built-in WebFetch/WebSearch with superior accuracy
- Clean markdown output optimized for LLM context windows
- JavaScript rendering and common block bypass
- Web, image, and news search with time/location filters
- Site mapping and URL discovery
- Parallel scraping with automatic rate limiting
- Search AND scrape in single operation
Changed
-
install.sh: Improved project root detection with 4-tier priority:
- Composer vendor install (
vendor/webconsulting/webconsulting-skills) - Subdirectory of project (
project/webconsulting-skills/) - Current working directory (if has
composer.json) - Script directory (fallback)
- Composer vendor install (
-
README.md: Updated installation instructions
- Split Quick Start into "Standalone" and "Clone as Subdirectory" options
- Added pre-pull steps (git stash) to Updating Skills section
- Added note about symlink recreation behavior
Added
-
marketing-skills: Marketing skills collection from Corey Haines
- CRO: Page conversion rate optimization framework
- Copywriting: Headlines, CTAs, page structure, copy formulas
- SEO: Technical and on-page SEO audit checklist
- Psychology: 70+ mental models for marketing and persuasion
- Pricing: Pricing strategy, tiers, value metrics, freemium vs trial
-
postgres-best-practices: Postgres performance optimization skill from Supabase
- Query performance: indexes, composite indexes, covering indexes, partial indexes
- Connection management: pooling, limits, idle timeouts
- Security & RLS: Row Level Security patterns, auth.uid() optimization, least privilege
- Schema design: data types, FK indexes, partitioning, primary key strategies
- Concurrency & locking: short transactions, deadlock prevention, SKIP LOCKED
- Data access patterns: batch inserts, N+1 elimination, cursor pagination, UPSERT
- Monitoring: pg_stat_statements, EXPLAIN ANALYZE, autovacuum tuning
- Advanced features: JSONB indexes, full-text search with tsvector
- SKILL-SUPABASE.md supplement: auth.uid(), Edge Functions, MCP server, Realtime
-
deepfake-detection: Multimodal media authentication and synthetic media forensics
- PRNU/PCE sensor fingerprint analysis
- IGH classification for GAN detection
- DQ probability maps for splicing detection
- Semantic forensics (shadows, reflections, physics)
- C2PA/CAI content provenance verification
- Voice clone detection patterns
- LLM-augmented sensemaking for post-empirical media
-
ai-search-optimization: New skill for AI search visibility (AEO/GEO)
- Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) strategies
- Schema.org markup (FAQPage, Article, HowTo, Organization) with sourced statistics
- E-E-A-T signals for AI trust and domain authority
- robots.txt configuration for AI crawlers (GPTBot, PerplexityBot, ClaudeBot)
- Content freshness strategies by platform (Perplexity, ChatGPT, Google AI)
- AI search visibility monitoring tools reference
- TYPO3 implementation with EXT:schema and EXT:ai_llms_txt (TYPO3 v14)
- Markdown/MDX implementation (Next.js, Astro) with JSON-LD components
- llms.txt specification and implementation guide
- Raw MDX view with URL parameters (.md suffix, ?raw=true)
3.0.0 - 2026-01-04
Added
- 11 new skills - Major expansion of skill library
context7: Library documentation lookup via Context7 REST APIcli-tools: CLI tool management and auto-installationsecurity-audit: Security audit patterns (OWASP, XXE, SQLi, XSS, CVSS)typo3-conformance: Extension standards compliance checkerenterprise-readiness: OpenSSF, SLSA, supply chain securitytypo3-testing: Unit, functional, E2E, architecture, mutation testingtypo3-extension-upgrade: Systematic extension upgrades (Rector, Fractor)typo3-docs: Documentation using docs.typo3.org standardsphp-modernization: PHP 8.x patterns, PHPStan level 10, DTOs, enumstypo3-core-contributions: TYPO3 Core contribution workflow (Gerrit, Forge)
Changed
- README.md: Updated documentation and attribution
- AGENTS.md: Reorganized with new skill categories, updated session profiles, expanded compatibility matrix
- typo3-ddev: Added multi-version testing section for extension development, container priority guidance
Attribution
Thanks to Netresearch DTT GmbH for their contributions.
2.0.0 - 2026-01-02
Changed
-
typo3-datahandler: Command examples now use correct
move/copysyntax. Clarified transaction guidance — DataHandler already wraps its own writes, outer transactions only needed when bundling adjacent operations on the same connection. Seeskills/typo3-datahandler/SKILL.md -
typo3-ddev: Snapshot deletion now uses
ddev snapshot delete. Added warning that "nuclear option" drops the database volume to prevent accidental data loss. Seeskills/typo3-ddev/SKILL.md -
typo3-seo: Robots.txt sitemap example now uses
{getEnv:TYPO3_SITE_URL}sitemap.xmlfor accurate URLs matching site base instead of request host. Seeskills/typo3-seo/SKILL.md -
typo3-security: Install Tool password guidance no longer shows empty placeholder. Now instructs setting hashed password per environment to prevent unsafe defaults. See
skills/typo3-security/SKILL.md
1.0.0 - 2026-01-01
Added
- Initial release with 7 agent skills
webconsulting-branding: Design tokens, MDX components, brand guidelinestypo3-datahandler: Transactional database operations via DataHandlertypo3-ddev: Local DDEV development environmenttypo3-seo: SEO configuration with EXT:seotypo3-security: Security hardening checklisttypo3-rector: TYPO3 upgrade patterns with Rectortypo3-update: TYPO3 v14 migration guide- Cursor IDE integration with master architect rule
- Installation script (
install.sh)
What's inside
4 release entries, 3 major versions, 30+ changed items, 20+ added skills, attribution lines
Change this for your project
- Replace
dirnbauer/webconsulting-skillswith your own repository path - Replace
dirnbauer/webconsulting-skillsin GitHub compare URLs with your own repo - Replace
webconsultingin skill names likewebconsulting-brandingwith your brand prefix
Where it goes
Keep in docs/ or alongside the feature. Agents read it to implement against a defined contract.
Worth borrowing
- Grouping changes by skill or component name for easy scanning
- Linking each change to its detailed SKILL.md file for deeper context
- Using Keep a Changelog format for consistent release notes across projects
Related Documents
GPU Selection Guide for Large Language Models (LLMs)
Guides GPU selection for LLM inference, fine-tuning, and training by mapping model sizes, precision levels, and budgets to VRAM requirements.
Community AI Agent Skills Discovery Sources
Catalogs 50+ platforms, repositories, directories, and communities for discovering and sharing AI agent skills across multiple coding tools.
ReleaseKit - Technical Requirements Document
Specifies a Go library and CLI for release automation with conventional commit parsing, validation checks, and workflow orchestration.
api_llm Specification
Defines a workspace of thin HTTP API clients for major LLM providers with no abstraction layer and explicit developer control.