davila7/claude-code-templates
306 skills132k installs9M starsGitHub
Install
npx skills add https://github.com/davila7/claude-code-templatesSkills in this repo
1Senior Data Scientistsenior-data-scientist is a davila7/claude-code-templates skill embedding experiment design, feature engineering, and production ML patterns for statistical modeling, causal inference, and advanced analytics work. Quick-start scripts include experiment_designer.py, feature_engineering_pipeline.py, and model_evaluation_suite.py for structured analysis workflows. Reference guides load on demand for statistical_methods_advanced.md, experiment_design_frameworks.md, and feature_engineering_patterns.md covering step-by-step processes, architecture patterns, and implementation examples. Production patterns span scalable distributed processing, ML model deployment with A/B testing infrastructure, feature store integration, drift detection, and high-throughput real-time inference with batching and auto-scaling. Best practices emphasize test-driven development, monitoring, automated deployments, canary releases, and team leadership standards for coding and cross-functional collaboration. Performance targets document P50 under fifty milliseconds, P95 under one hundred milliseconds, P99 under two hundred milliseconds, over one thousand requests per second throughput, and 99.9 percent uptime av.2.9kinstalls2Senior BackendThe senior-backend skill is a comprehensive backend toolkit covering API scaffolding, database migration analysis, load testing, and security reference guides for Node.js, Express, Go, Python, GraphQL, REST, and PostgreSQL stacks. Three Python scripts provide automated api scaffolder, database migration tool, and api load tester workflows with configurable templates and quality checks. Reference docs cover API design patterns, database optimization, and backend security practices with anti-patterns and troubleshooting sections. The documented tech stack spans TypeScript, JavaScript, Python, Go, React, Next.js, Prisma, NeonDB, Supabase, Docker, Kubernetes, Terraform, and major clouds. Development workflow guidance walks through dependency install, environment configuration, analyzer runs, and deployment commands. Best practices stress input validation, parameterized queries, authentication, caching, monitoring, and maintainable naming. Agents invoke it when designing APIs, optimizing queries, implementing auth, reviewing backend code, or running analysis scripts against a project path.2.4kinstalls3Excel AnalysisRead Excel file df pd read_excel data xlsx sheet_name Sheet1 Display first few rows print df head Basic statistics print df describe Read all sheets excel_file pd ExcelFile workbook xlsx for sheet_name in excel_file sheet_names df pd read_excel excel_file sheet_name sheet_name print f n sheet_name print df head The Excel Analysis agent skill provides documented workflows prerequisites triggers and safety guidance from its SKILL md source Agents load it when user requests match the description and follow step by step instructions without inventing capabilities It integrates with standard agent tooling for the tasks inputs outputs and failure modes described in the repository documentation2.2kinstalls4Senior FrontendComprehensive frontend development skill for building modern, performant web applications using ReactJS, NextJS, TypeScript, Tailwind CSS. Includes component scaffolding, performance optimization, bundle analysis, and UI best practices. Use when developing frontend features, optimizing performance, implementing UI/UX designs, managing state, or reviewing frontend code. The senior-frontend skill documents workflows and patterns from the repository SKILL.md. --- name: senior-frontend description: Comprehensive frontend development skill for building modern, performant web applications using ReactJS, NextJS, TypeScript, Tailwind CSS. Includes component scaffolding, performance optimization, bundle analysis, and UI best practices. Use when developing frontend features, optimizing performance, implementing UI/UX designs, managing state, or reviewing frontend code. --- # Senior Frontend Complete toolkit for senior frontend with modern tools and best practices. ## Quick Start ### Main Capabilities This skill provides three core capabilities through automated scripts: ```bash # Script 1: Component Generator python scripts/component_generator.py [options] # Script 2: Bundle Analyzer python.1.8kinstalls5Literature ReviewConduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.). This skill should be used when conducting systematic literature reviews, meta-analyses, research synthesis, or comprehensive literature searches across biomedical, scientific, and technical domains. Creates professionally formatted markdown documents and PDFs wit The literature-review skill documents workflows and patterns from the repository SKILL.md. --- name: literature-review description: Conduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.). This skill should be used when conducting systematic literature reviews, meta-analyses, research synthesis, or comprehensive literature searches across biomedical, scientific, and technical domains. Creates professionally formatted markdown documents and PDFs with verified citations in multiple citation styles (APA, Nature, Vancouver, etc.). allowed-tools: [Read, Write, Edit, Bash] --- # Literature Review ## Overview Conduct systematic, comprehensive literature reviews following rigorous academic methodology. Search multipl.1.8kinstalls6MarkitdownThe markitdown skill wraps Microsoft's MarkItDown Python tool for converting documents to token-efficient Markdown for LLM processing. Supported formats include PDF, DOCX, PPTX, XLSX, images with OCR, audio with transcription, HTML, CSV, JSON, XML, ZIP, YouTube URLs, and EPUB. Optional AI-enhanced image descriptions and scientific-schematics integration suggest generating publication diagrams via scripts/generate_schematic.py for complex workflows. Allowed tools: Read, Write, Edit, Bash. Use when users need file-to-markdown conversion, document ingestion for RAG, or office format preprocessing before agent analysis. Not a general document editor; focused on conversion pipeline to clean structured Markdown.1.7kinstalls7Scientific WritingThe scientific-writing skill produces research manuscripts as flowing paragraphs never bullet points in final output, using a two-stage process of section outlines with research-lookup then conversion to prose. Covers IMRAD structure, APA AMA Vancouver Chicago IEEE citations, figures and tables, and reporting guidelines including CONSORT STROBE and PRISMA. Mandatory visual enhancement requires at least one AI-generated figure via scientific-schematics per paper, preferably two to three for comprehensive work. Applies to abstracts, introductions, methods, results, discussion, journal formatting, reviewer response, and field-specific nomenclature. Critical principle: always write full paragraphs; use outlines only as intermediate step. Invoke when drafting research papers, revising journal submissions, or formatting citations for academic publication workflows. Two-stage: research-lookup outlines then flowing IMRAD prose. Never bullet points in final manuscript paragraphs. Citations: APA, AMA, Vancouver, Chicago, IEEE styles.1.7kinstalls8Market Research ReportsThe market-research-reports skill produces professional-grade strategic documents of fifty plus pages modeled after McKinsey, BCG, Gartner, and Forrester deliverables. Key features include visual-rich content with five to six key diagrams at start, deep research-lookup integration for market data, multi-framework analysis using Porter Five Forces, PESTLE, SWOT, BCG Matrix, and TAM/SAM/SOM sizing, plus LaTeX output compiled to PDF via market_research.sty styling. Visual tools include scientific-schematics for growth trajectories and competitive matrices and generate-image for supporting graphics. Use cases span investment decisions, industry reports, competitive landscapes, market entry, M&A due diligence, thought leadership, go-to-market strategy, regulatory impact analysis, and new product business cases. Invoke when users need comprehensive market analysis with actionable recommendations and implementation roadmaps. 50+ page consulting-style reports in LaTeX to PDF. Frameworks: Porter, PESTLE, SWOT, BCG Matrix, TAM/SAM/SOM.1.6kinstalls9Exploratory Data AnalysisThe exploratory data analysis skill Perform comprehensive exploratory data analysis on scientific data files across 200+ file formats. This skill should be used when analyzing any scientific data file to understand its structure, content, quality, and characteristics. Automatically detects file type and generates detailed markdown reports with format-specific analysis, quality metrics, and downstream analysis recommendations. Covers chemistry, bioinformatics, microscopy, spectroscopy, proteomics, metabolomics, and general scientific data formats. Documentation covers workflows, commands, and guardrails agents should follow when users invoke this capability. Key documented areas include Automatic detection and analysis of 200+ scientific file formats; Comprehensive format-specific metadata extraction; Data quality and integrity assessment; Statistical summaries and distributions. Reference commands include User: "Analyze data.fastq"; → Extension: .fastq. Use when developers or agents need structured guidance for exploratory data analysis tasks with evidence grounded in the bundled SKILL.md rather than generic advice.1.6kinstalls10Mobile DesignMobile Design System teaches touch-first principles, platform conventions, and performance patterns for iOS and Android development. Developers use it when building React Native, Flutter, or native mobile apps to ensure UX respects mobile constraints (limited screen, imprecise touch, battery, offline). Core workflows include validating touch targets and gesture handling via the mobile_audit.py script, reading platform-specific design files (platform-ios.md, platform-android.md) before starting work, applying Fitts' Law principles (44-48px targets, thumb-zone CTAs), and avoiding AI defaults like ScrollView lists and AsyncStorage tokens. Includes mandatory checkpoint forcing explicit platform/framework/anti-pattern declarations before coding.1.5kinstalls11Senior Data EngineerSenior Data Engineer skill provides production-grade expertise in building scalable data pipelines, ETL/ELT systems, and ML infrastructure. Covers data modeling, pipeline orchestration with Airflow and dbt, real-time processing via Kafka, and distributed computing on Spark. Developers use this when designing data architectures, optimizing workflows across PostgreSQL/BigQuery/Snowflake, implementing data quality validation, and deploying ML models with monitoring via MLflow and Prometheus. Includes patterns for horizontal scaling, fault tolerance, A/B testing, and automated retraining with security and compliance built-in.1.5kinstalls12Stable Diffusion Image Generationstable-diffusion-image-generation is an agent skill from davila7/claude-code-templates that state-of-the-art text-to-image generation with stable diffusion models via huggingface diffusers. use when generating images from text prompts, performing image-to-image translation, inpainting, or bu. # Stable Diffusion Image Generation Comprehensive guide to generating images with Stable Diffusion using the HuggingFace Diffusers library. ## When to use Stable Diffusion **Use Stable Diffusion when:** - Generating images from text descriptions - Performing image-to-image translation (style transfer, enhancement) - Inpainting (filling in masked Developers invoke stable-diffusion-image-generation during build/integrations work for ai & agent building tasks. The skill documents triggers, prerequisites, and step-by-step workflows grounded in SKILL.md. Compatible with Claude Code, Cursor, and Codex agent runtimes that load marketplace skills. Review the Security Audits panel on this listing before installing in production environments.1.5kinstalls13Ui Design SystemThe ui-design-system skill is a toolkit for senior UI designers creating and maintaining scalable design systems. Core capabilities include design token generation for colors, typography, and spacing, component system architecture, responsive design calculations, accessibility compliance, and developer handoff documentation. The design_token_generator.py script produces complete token sets from a brand color with modern, classic, or playful styles exported as JSON, CSS, or SCSS. Generated outputs cover color palettes, modular typography scales, an eight-point spacing grid, shadow and animation tokens, and responsive breakpoints. Use when developers or designers need consistent visual tokens, component documentation structure, or handoff files that bridge design and implementation teams. The skill focuses on systematic token generation rather than reviewing existing component code.1.4kinstalls14Ui Ux Pro Maxui-ux-pro-max is an agent skill from davila7/claude-code-templates that ui/ux design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 9 stacks (react, next.js, vue, svelte, swiftui, react native, flutter, tailwind, shadcn/ui). actions: plan, build, creat. # UI/UX Pro Max - Design Intelligence Comprehensive design guide for web and mobile applications. Contains 50+ styles, 97 color palettes, 57 font pairings, 99 UX guidelines, and 25 chart types across 9 technology stacks. Searchable database with priority-based recommendations. ## When to Apply Reference these guidelines when: - Designing new UI Developers invoke ui-ux-pro-max during operate/infra work for cloud & infrastructure tasks. The skill documents triggers, prerequisites, and step-by-step workflows grounded in SKILL.md. Compatible with Claude Code, Cursor, and Codex agent runtimes that load marketplace skills.1.4kinstalls15Scientific VisualizationThe scientific-visualization skill is designed for create publication figures with matplotlib/seaborn/plotly. Multi-panel layouts, error bars, significance markers, colorblind-safe, export PDF/EPS/TIFF, for journal-ready. Scientific Visualization Overview Scientific visualization transforms data into clear, accurate figures for publication. Create journal-ready plots with multi-panel layouts, error bars, significance markers, and colorblind-safe palettes. Invoke when the user asks about scientific visualization or related SKILL.md workflows.1.3kinstalls16WhisperThe whisper skill is designed for openAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M. Whisper - Robust Speech Recognition OpenAI's multilingual speech recognition model. Invoke when the user asks about whisper or related SKILL.md workflows.1.2kinstalls17Generate ImageThe generate-image skill generate or edit images using AI models FLUX Gemini Use for general-purpose image generation including photos illustrations artwork visual assets concept art and any image that isn't a technical diagram or schematic For flowcharts circuits pathways and technical diagrams use the scientific-schematics skill instead Generate Image Generate and edit high-quality images using OpenRouter's image generation models including FLUX 2 Pro and Gemini 3 Pro When to Use This Skill Use generate-image for Photos and photorealistic images Artistic illustrations and artwork Concept art and visual concepts Visual assets for presentations or documents Image editing and modifications Any general-purpose image generation needs Use scientific-schematics instead for Flowcharts and process diagrams Circuit diagrams and electrical schematics Biological pathways and signaling cascades System architecture diagrams CONSORT diagrams and methodology flowcharts Any technical schematic diagrams Quick Start Use the scripts generate_image py script to generate or edit images bash Generate a new image python scripts generate_image py A beautiful sunset over mountains Edit an existing image.1.2kinstalls18Scientific SlidesThe scientific-slides skill build slide decks and presentations for research talks Use this for making PowerPoint slides conference presentations seminar talks research presentations thesis defense slides or any scientific talk Provides slide structure design templates timing guidance and visual validation Works with PowerPoint and LaTeX Beamer Scientific Slides Overview Scientific presentations are a critical medium for communicating research sharing findings and engaging with academic and professional audiences This skill provides comprehensive guidance for creating effective scientific presentations from structure and content development to visual design and delivery preparation Key Focus Oral presentations for conferences seminars defenses and professional talks CRITICAL DESIGN PHILOSOPHY Scientific presentations should be VISUALLY ENGAGING and RESEARCH-BACKED Avoid dry text-heavy slides at all costs Great scientific presentations combine Compelling visuals High-quality figures images diagrams not just bullet points Research context Proper citations from research-lookup establishing credibility Minimal text Bullet points as prompts YOU provide the explanation verbally Professi.1.2kinstalls19Citation ManagementThe citation-management skill comprehensive citation management for academic research Search Google Scholar and PubMed for papers extract accurate metadata validate citations and generate properly formatted BibTeX entries This skill should be used when you need to find papers verify citation information convert DOIs to BibTeX or ensure reference accuracy in scientific writing Citation Management Overview Manage citations systematically throughout the research and writing process This skill provides tools and strategies for searching academic databases Google Scholar PubMed extracting accurate metadata from multiple sources CrossRef PubMed arXiv validating citation information and generating properly formatted BibTeX entries Critical for maintaining citation accuracy avoiding reference errors and ensuring reproducible research Integrates seamlessly with the literature-review skill for comprehensive research workflows When to Use This Skill Use this skill when Searching for specific papers on Google Scholar or PubMed Converting DOIs PMIDs or arXiv IDs to properly formatted BibTeX Extracting complete metadata for citations authors title journal year etc Validating existing citations.1.2kinstalls20Marketing Strategy PmmThe marketing-strategy-pmm skill product marketing, positioning, GTM strategy, and competitive intelligence. Includes ICP definition, April Dunford positioning methodology, launch playbooks, competitive battlecards, and international market entry guides. Use when developing positioning, planning product launches, creating messaging, analyzing competitors, entering new markets, enabling sales, or when user mentions product marketing, positioning, GTM, go-to-market, competitive analysis, market entry, or sales enablement. # Marketing Strategy & Product Marketing Expert Product Marketing playbook for Series A+ startups expanding internationally with hybrid PLG/Sales-Led motion. ## Keywords product marketing, positioning, GTM, go-to-market strategy, competitive analysis, competitive intelligence, battlecards, ICP, ideal customer profile, messaging, value proposition, product launch, market entry, international expansion, sales enablement, win loss analysis, PMM, product marketing manager, market positioning, competitive landscape, sales training ## Role Coverage This skill serves: - **Product Marketing Manager (PMM)** - Positioning, messaging, competitive intel, launches - **Head of M.1.2kinstalls21RemotionThe remotion skill best practices and comprehensive guide for Remotion programmatic video creation in React with animations compositions and media handling Remotion Video Creation in React Comprehensive skill set for creating programmatic videos using Remotion a framework for creating videos programmatically using React When to use Use this skill whenever you are dealing with Remotion code to obtain domain-specific knowledge about Creating video compositions with React components Animating elements using frame-based animations Working with audio video and image assets Building charts and data visualizations Implementing text animations and captions Using 3D content with Three js Applying transitions and sequencing Integrating Tailwind CSS and Lottie animations Core Concepts Remotion allows you to create videos using React Components Build video content with familiar React syntax Frame-based Animations All animations driven by useCurrentFrame hook Compositions Define video compositions with duration dimensions and props Assets Import and manipulate images videos audio and fonts Rendering Export videos programmatically with customizable settings Key Features Frame-by-frame control o.1.2kinstalls22Senior DevopsThe senior-devops skill comprehensive DevOps skill for CI/CD, infrastructure automation, containerization, and cloud platforms (AWS, GCP, Azure). Includes pipeline setup, infrastructure as code, deployment automation, and monitoring. Use when setting up pipelines, deploying applications, managing infrastructure, implementing monitoring, or optimizing deployment processes. # Senior Devops Complete toolkit for senior devops with modern tools and best practices. ## Quick Start ### Main Capabilities This skill provides three core capabilities through automated scripts: ```bash # Script 1: Pipeline Generator python scripts/pipeline_generator.py [options] # Script 2: Terraform Scaffolder python scripts/terraform_scaffolder.py [options] # Script 3: Deployment Manager python scripts/deployment_manager.py [options] ``` ## Core Capabilities ### 1. Pipeline Generator Automated tool for pipeline generator tasks. **Features:** - Automated scaffolding - Best practices built-in - Configurable templates - Quality checks **Usage:** ```bash python scripts/pipeline_generator.py <project-path> [options] ``` ### 2. Terraform Scaffolder Comprehensive analysis and optimization tool. **Features:** - Deep.1.2kinstalls23Prompt EngineerThe prompt-engineer skill from claude-code-templates teaches structured prompt design for Claude Code agents including role definition, constraint blocks, few-shot examples, and evaluation loops. It covers decomposing vague user goals into testable instructions, separating system behavior from tool usage rules, and iterating prompts against failure cases. Agents apply templates for coding assistants, reviewers, and workflow orchestrators while avoiding ambiguous or conflicting instruction stacks. The skill emphasizes measurable success criteria and concise prompt sections that models can follow reliably during long multi-step sessions.1.1kinstalls24Mermaid Diagram SpecialistThe mermaid-diagram-specialist skill mermaid diagram specialist for creating flowcharts, sequence diagrams, ERDs, # Mermaid Diagram Specialist ## Overview **Purpose**: Expert in creating comprehensive Mermaid diagrams for documentation, architecture visualization, and process mapping **Category**: Tech **Primary Users**: tech-writer, architecture-validator, product-technical, tech-lead ## When to Use This Skill - Creating architecture documentation - Visualizing workflows and processes - Documenting data models (ERDs) - Explaining sequence flows - Creating state machines - Documenting component relationships - Creating decision trees - Visualizing user journeys ## Prerequisites **Required:** - Understanding of the system/process to document - Access to technical specifications - Knowledge of diagram type needed **Optional:** - Design system colors for consistency - Existing documentation to reference ## Input **What the skill needs:** - Process/system description - Entities and relationships (for ERDs) - Component interactions (for sequence diagrams) - Architecture layers (for C4 diagrams) - States and transitions (for state diagrams) ## Workflow ### Step 1: Diagram Type Selection.1.1kinstalls25Ux Researcher DesignerThe ux-researcher-designer skill is a Senior UX Designer and Researcher toolkit for user-centered discovery and validation. Core capabilities include data-driven persona generation, customer journey mapping, usability testing frameworks, research synthesis, and design validation methods. The persona_generator.py script analyzes user behavior patterns, identifies archetypes, extracts psychographics, generates scenarios, and scores confidence from sample size when given JSON input. Teams invoke it for persona creation, journey mapping, usability testing design, and translating research into design implications. The skill focuses on structured research outputs rather than visual UI implementation. It expects behavioral or interview data as input for persona scripts and supports repeatable research workflows inside agent sessions.1.1kinstalls26Senior Qasenior-qa is a comprehensive QA skill from davila7/claude-code-templates aimed at ReactJS, NextJS, and NodeJS codebases. It provides three automated Python scripts—a test suite generator, a coverage analyzer, and an E2E testing setup tool—plus guidance on test strategy, manual testing, and quality metrics. Developers reach for senior-qa when designing test plans, writing test cases, implementing automation, or closing coverage gaps before release. The skill covers unit, integration, and end-to-end patterns with modern tooling conventions, making it useful both for greenfield test scaffolding and for hardening existing apps that lack systematic QA coverage.1kinstalls27Jirajira is a natural-language Jira skill for coding agents that auto-detects the best backend on startup. It runs `which jira` for the CLI backend, falls back to mcp__atlassian__* MCP tools when available, and guides setup when neither exists. Triggers include issue keys like PROJ-123, plus keywords jira, ticket, sprint, backlog, and issue. Developers reach for jira when they want to create, view, or update issues, check sprint status, or manage workflow state from Claude Code without switching to the Jira web UI.1kinstalls28Scientific Schematicsscientific-schematics from davila7/claude-code-templates provides AI-powered generation of publication-ready scientific diagrams, CONSORT flowcharts, and research schematics. Example usage scripts require Python 3.10 or newer, the requests package, and an OPENROUTER_API_KEY environment variable pointing to openrouter.ai. The bundled bash examples demonstrate end-to-end schematic generation workflows developers can adapt in research repositories. Developers reach for scientific-schematics when they need agent-assisted research figures instead of manual Illustrator or draw.io sessions for papers, protocols, or trial documentation.1kinstalls29Video DownloaderVideo Downloader is a Claude Code Templates skill that fetches videos from YouTube and other platforms directly to local storage with selectable quality and format options. Developers use it to save educational content, conference talks, webinars, or their own published videos for offline viewing, archival, or downstream editing and repurposing workflows. The skill handles platform fetching and quality selection so agents do not require manual yt-dlp command assembly for common download tasks. It fits content research, documentation, and media preparation phases where source video files are needed locally rather than streamed links.993installs30Matplotlibmatplotlib is a Claude Code Templates skill that packages a quick reference for Matplotlib's core plotting API—Figure creation, `plt.subplots` grids, `add_subplot`, `add_axes`, `savefig` with `dpi` and `bbox_inches='tight'`, `tight_layout`, and `suptitle`. Developers reach for matplotlib when agents must produce bar, line, or multi-panel figures in notebooks, ETL reports, or backend analytics scripts and syntax mistakes would waste iteration cycles. The reference shows concrete patterns like `fig, ax = plt.subplots(nrows=1, ncols=1, figsize=(10, 6))` and `fig.savefig(filename, dpi=300, bbox_inches='tight')`. It is documentation-shaped guidance meant to keep generated Python visualization code accurate on first pass.982installs31Google Analyticsgoogle-analytics is an analytics integration skill from davila7/claude-code-templates that wires coding agents to Google Analytics 4 using a numeric Property ID and a Google Cloud service account JSON key via GOOGLE_APPLICATION_CREDENTIALS. Setup requires creating a service account in Google Cloud Console, storing the JSON key outside the git repository, and configuring GOOGLE_ANALYTICS_PROPERTY_ID from GA4 Admin → Property Settings. Developers reach for google-analytics when they want agents to pull real traffic, conversion, and engagement metrics into automated reports instead of manually exporting GA4 dashboards. The skill targets teams already running GA4 on a live property who need programmatic access for recurring performance summaries.978installs32Senior Architectsenior-architect is a Claude Code reference skill from davila7/claude-code-templates that walks developers through architecture patterns with TypeScript implementation snippets, when-to-use scenarios, benefits, and explicit trade-offs. The guide covers best-practice implementations and advanced techniques so agents can recommend component boundaries, interface contracts, and scalability choices instead of ad hoc structure. Developers reach for senior-architect when scoping a new microservice, refactoring a monolith, or reviewing whether a proposed design repeats known anti-patterns. The skill frames each pattern with concrete code examples and decision criteria rather than generic high-level advice.978installs33Senior Computer Visionsenior-computer-vision is a Claude Code Templates skill for world-class computer vision architecture aimed at senior engineers. It encodes production-first design—planning for 10x load growth, 99.9% uptime targets, maintainable modules, and full observability—plus performance patterns like batch processing, caching, and resource-aware algorithms. Security and privacy guardrails cover input validation, encryption, access control, and audit logging. Developers reach for senior-computer-vision when drafting distributed CV pipelines, inference services, or enterprise-scale vision platforms instead of notebook-only prototypes.976installs34Scientific Critical Thinkingscientific-critical-thinking is a Claude Code Templates skill that systematically counters confirmation bias, hindsight bias, and publication bias when planning, researching, or evaluating AI-powered products and features. The reference material documents how researchers skew studies toward preexisting beliefs, reinterpret ambiguous results as supportive, and overestimate predictability after outcomes are known. Mitigation patterns include preregistering hypotheses, actively seeking disconfirming evidence, blinded analysis, and considering alternative hypotheses. Developers reach for scientific-critical-thinking when designing experiments, reviewing literature, or making product decisions where biased reasoning could steer AI feature investments toward false confidence.972installs35Senior FullstackSenior-fullstack is a Claude Code skill from davila7/claude-code-templates that acts as a senior-engineer playbook for full-stack architecture. It catalogs pattern implementations with TypeScript examples, when-to-use scenarios, benefits, and trade-offs across code organization, performance, and security practices. Agents apply it while scaffolding services, reviewing module boundaries, or aligning new features with established layering and best-practice conventions. Reach for senior-fullstack when you want agent-generated code to follow senior-level structure instead of ad-hoc file layouts, or when refactoring a growing codebase toward clearer separation of concerns and safer defaults.959installs36Muimui is a comprehensive skill from davila7/claude-code-templates for building React applications with Material-UI v7, released March 2025. It covers component usage, the sx prop styling system, theme integration, responsive design, and MUI-specific utilities with type-safe patterns. Developers reach for mui when styling with sx, migrating from MUI v6 breaking changes, wiring themes, or composing accessible layouts with one of the most widely used React UI libraries. The skill helps agents apply consistent MUI v7 conventions instead of mixing deprecated APIs or ad hoc CSS. Use mui during feature UI work, design-system alignment, or refactors touching MUI imports and theme tokens.951installs37Ceo Advisorceo-advisor is a strategic advisory skill from davila7/claude-code-templates with 795 installs listed on skills.sh. It gives developers and technical leads CEO-style framing when they must weigh scope cuts, positioning choices, and go-to-market tradeoffs before writing code. The skill does not generate application code; it structures prioritization conversations so engineering effort aligns with business constraints. Reach for ceo-advisor at project kickoff, pivot discussions, or when a feature backlog needs ruthless prioritization against positioning and launch timing. It complements technical planning skills by focusing on executive-level tradeoff analysis rather than implementation detail.938installs38Seo Optimizerseo optimizer is a Search Engine Optimization agent skill from davila7/claude-code-templates that helps developers improve organic search visibility through content strategy, technical SEO, and keyword research. The skill covers on-page optimization, meta tag authoring, schema markup implementation, Core Web Vitals remediation, and ranking analysis so public websites ship with search-engine-ready structure. Developers reach for seo optimizer when a site launches or stagnates in rankings and they need agent-guided audits of titles, headings, internal links, structured data, and performance signals without hiring a separate SEO specialist. Guidance spans strategic planning—keyword targeting and content gaps—and tactical fixes such as canonical tags, Open Graph fields, JSON-LD schemas, and Lighthouse-driven Core Web Vitals improvements. Use seo optimizer during launch prep, content refreshes, or technical SEO debugging when organic traffic is a primary distribution channel for a docs site, SaaS marketing page, or developer blog.916installs39Scientific Brainstormingscientific-brainstorming is a davila7/claude-code-templates reference skill that applies structured brainstorming frameworks when standard prompting yields repetitive or shallow scientific ideas. The skill documents methodologies including SCAMPER, which provides seven approaches—Substitute, Combine, Adapt, Modify, Put to another use, Eliminate, and Reverse—for improving methods or adapting known techniques. Developers and research engineers invoke scientific-brainstorming during early hypothesis design, grant ideation, or experiment planning when they need disciplined ideation beyond generic chat brainstorming.908installs40Senior Prompt Engineersenior-prompt-engineer is an advanced skill for engineers designing production agentic systems who need prompts and architectures that scale beyond prototype chatbots. The skill encodes production-first principles: plan for 10x load, target 99.9% uptime, maintain clear documented interfaces, and instrument observability from day one. It covers performance-by-design patterns—efficient algorithms, resource awareness, strategic caching, and batch processing—plus security and privacy controls including input validation, encryption, access control, and audit logging. Advanced sections address distributed processing and enterprise-scale fault tolerance for multi-step agent pipelines. Reach for senior-prompt-engineer when defining system prompts, tool policies, or multi-agent roles before coding orchestration frameworks like CrewAI or LangGraph. The skill emphasizes reliable first-run behavior for customer-facing agents rather than one-off prompt hacks, making it suitable for staff-level prompt and agent architecture reviews inside Claude Code or Cursor.903installs41Clean Codeclean-code is a CRITICAL-priority Claude Code skill (version 2.0) from davila7/claude-code-templates that keeps agent-generated codebases concise, readable, and free of over-engineering. It codifies SRP, DRY, KISS, YAGNI, and Boy Scout principles with explicit rules against unnecessary comments and bloated abstractions. Allowed tools are Read, Write, and Edit so the skill shapes code as agents produce it. Developers invoke clean-code when AI output drifts verbose, over-abstracted, or comment-heavy and they want consistent pragmatic standards across functions and classes.901installs42Crewai Multi Agentcrewai-multi-agent is a development skill for Python engineers who outgrow simple sequential CrewAI crews and need event-driven orchestration with CrewAI Flows. The skill explains when to choose Flows over Crews: conditional branching, complex shared state, event listeners, and hybrid patterns that embed Crews inside Flow steps. It documents Flow basics with Python decorators including @start, @listen, @router, and logical combinators or_ and and_, plus Pydantic state models for typed execution paths. Reach for crewai-multi-agent when building approval gates, multi-stage pipelines with forks, or reactive agent workflows that must resume after external events. The skill contrasts sequential Crew collaboration with Flow-controlled routing so you pick the lighter Crew abstraction only when branching and state are unnecessary. It assumes an existing CrewAI Python environment and focuses on orchestration design rather than LLM prompt copy.895installs43Conversation Memoryconversation-memory is a davila7 claude-code-templates skill sourced from vibeship-spawner-skills under Apache 2.0 that specializes in memory systems for AI assistants. The skill guides retrieval, relevance scoring, and context surfacing so agents remember users across months without flooding the prompt with every past message. It distinguishes short-term working memory, long-term durable storage, and entity-based memory for people, projects, and concepts. Developers reach for conversation-memory when building Claude or Cursor agents that need selective recall, intentional forgetting, and memory persistence beyond a single chat window.893installs443d Web Experience3d-web-experience is an expert agent skill sourced from vibeship-spawner-skills under Apache 2.0 that guides developers through building interactive 3D scenes in the browser. The skill covers Three.js, React Three Fiber, Spline imports, and raw WebGL patterns for product configurators, 3D portfolios, and immersive marketing sites. It emphasizes when 3D adds real value versus visual noise, balancing visual impact with frame-rate budgets, keyboard and screen-reader accessibility, and progressive enhancement for users unfamiliar with 3D navigation. Developers reach for 3d-web-experience when scaffolding a new R3F canvas, choosing between Spline embeds and custom shaders, or auditing an existing scene for jank, layout shift, and mobile GPU limits before shipping a public-facing experience.884installs45Agile Product Owneragile-product-owner is a skill from davila7/claude-code-templates that includes a Python UserStoryGenerator class for INVEST-compliant user stories with acceptance criteria. Built-in personas cover end users (efficiency, simplicity, reliability, speed) and administrators (control, visibility, security, configuration). The skill helps developers and tech leads convert vague feature requests into well-formed stories agents can implement without scope drift. Reach for it when kicking off a sprint, breaking epics into backlog items, or defining testable acceptance criteria before coding starts.884installs46Quality Documentation Managerquality-documentation-manager is a Claude Code Templates skill for development teams who need documentation that tracks code changes instead of drifting stale. The skill generates reference materials, applies quality standards to docs assets, and enforces consistency rules as repositories evolve. Bundled asset files such as templates, boilerplate directories, images, and fonts support output the agent produces rather than loading large binaries into context. Reach for quality-documentation-manager when onboarding docs for a new module, refreshing API references after refactors, or auditing whether README and internal guides still match the codebase. The skill targets sustained documentation hygiene rather than one-off copywriting.884installs47Senior Securitysenior-security is a reference skill from davila7/claude-code-templates focused on cryptography implementation and senior-level secure coding practices. It documents patterns with descriptions, when-to-use scenarios, TypeScript implementation examples, benefits, and trade-offs so agents apply vetted approaches instead of inventing crypto primitives. Developers reach for senior-security when building login flows, token handling, encryption at rest or in transit, or any sensitive operation where a mistake is costly. The skill emphasizes best-practice implementations and flags common anti-patterns across multiple pattern sections. It complements llm-security, which targets OWASP LLM threats, by focusing on general application cryptography and secure engineering discipline.880installs48Langchainlangchain is a skill from davila7/claude-code-templates with a complete LangChain Agents Guide covering the ReAct pattern: reasoning, acting, observing tool results, and looping until tasks complete. Examples use langchain.agents.create_agent with ChatAnthropic and custom tools such as a calculator and search helper. The guide walks through basic agent creation, tool-calling setup, and streaming responses for interactive UIs. Developers reach for langchain when bootstrapping Python agents that combine Claude with callable tools instead of writing orchestration from scratch.879installs49Plotlyplotly is a concise recipe skill for Plotly Express and graph_objects bundled in the davila7 claude-code-templates collection. It organizes chart types by category with copy-ready Python snippets for scatter, line, bar, pie, and related visualizations, including options like trendlines, stacked bars, grouped bars, and horizontal orientation. Developers reach for plotly when a coding agent must pick the correct chart API on the first attempt for notebooks, Streamlit apps, or reporting pipelines. The readme focuses on practical px and graph_objects calls rather than abstract visualization theory. Use it when you need fast, correct Plotly syntax for exploratory analysis, dashboard widgets, or exported HTML figures without digging through scattered documentation.879installs50Code Reviewercode-reviewer is a davila7 Claude Code Templates skill that provides automated pull-request review across TypeScript, JavaScript, Python, Swift, Kotlin, and Go. Three core Python scripts—pr_analyzer.py and code_quality_checker.py among them—drive PR analysis, best-practice validation, security scanning, and checklist generation. Developers invoke it when reviewing GitHub pull requests, surfacing issues, or standardizing feedback before merge. The skill bundles modern linting and security patterns into agent-guided review workflows rather than ad-hoc comments.864installs51Xlsxxlsx from davila7/claude-code-templates is an Anthropic-origin Claude Code skill (© 2025 Anthropic, PBC) that lets a coding agent read, create, and update Excel spreadsheets directly from natural language instructions. The template packages Anthropic's official xlsx skill materials under Anthropic Consumer or Commercial Terms of Service with additional usage restrictions documented in the license header. Developers invoke xlsx when spreadsheet files are primary inputs or outputs—financial models, operational reports, data exports, or bulk tabular updates—without switching to manual Excel editing mid-session. The skill fits agent workflows where .xlsx deliverables must be produced or modified programmatically from chat instructions. With 749 catalog installs, developers reach for xlsx when tabular files are the main artifact—not when the task is PDF generation, plain CSV one-liners without Excel features, or database schema design without spreadsheet deliverables.864installs52Qdrant Vector SearchQdrant Vector Search is an advanced usage guide skill that helps developers stand up Qdrant for semantic search and agent vector memory. The skill covers distributed deployment with Raft consensus, docker-compose cluster topology, HTTP and gRPC port mapping, and sharding configuration patterns. The readme excerpt documents a 3-node Qdrant cluster with per-node storage volumes and cluster environment variables such as QDRANT__CLUSTER__ENABLED. Developers reach for Qdrant Vector Search when embedding pipelines are ready but production cluster YAML, sharding rules, and multi-node coordination still need correct defaults. The skill targets agent builders and backend engineers integrating Qdrant instead of guessing distributed vector-database settings.860installs53Scikit Learnscikit-learn is a model selection and evaluation reference skill from davila7/claude-code-templates that guides AI coding agents through scikit-learn workflows. It covers train-test splits (default 75/25), stratified splits, three-way train/validation/test splits, cross-validation, hyperparameter search, and model comparison patterns with runnable Python examples. Developers reach for scikit-learn when they need sklearn.model_selection APIs wired correctly instead of guessing split ratios or evaluation metrics. The skill emphasizes reproducible splits with random_state and class-preserving stratify options. It fits mid-build ML features where evaluation rigor determines which model ships to production inference code.860installs54Frontend Designfrontend-design is an agent skill distributed through davila7/claude-code-templates that guides Claude to produce distinctive, production-grade frontend interfaces and component structures rather than generic AI-generated layouts. The skill emphasizes intentional typography, color, motion, and spatial composition to avoid converging on common default aesthetics. Developers reach for frontend-design when scaffolding new web UIs, redesigning components, or requesting landing pages and dashboards that must look professionally designed. The package is Apache-2.0 licensed and ships as part of the widely used claude-code-templates collection referenced across Claude Code workflows.851installs55Pptxpptx is a davila7/claude-code-templates skill for template-based PowerPoint editing and generation. The workflow analyzes an existing template.pptx with python scripts/thumbnail.py and python -m markitdown, reviews thumbnails.jpg for layouts, then maps each content section to varied slide types—multi-column, image-plus-text, full-bleed, and callout layouts instead of repetitive title-plus-bullet slides. Developers reach for pptx when turning architecture notes, sprint reviews, or technical briefs into polished .pptx decks without manually rebuilding layouts in PowerPoint.842installs56Mermaid Diagramsmermaid-diagrams is a comprehensive agent skill for creating professional software diagrams using Mermaid's text-based syntax directly from natural language prompts. The skill covers system architecture, code structure, database models, user flows, and component diagrams that render from simple text definitions and stay version-controlled alongside source code. Diagrams update easily in pull requests because they are plain text, not binary image exports. Developers reach for mermaid-diagrams when they need to visualize C4 contexts, ER models, sequence flows, or system designs without opening a separate diagramming tool.841installs57Docker Expertdocker-expert is a proactive Claude skill embodying deep Docker containerization knowledge: multi-stage builds, image size reduction, security hardening, networking, and Docker Compose multi-service orchestration. Developers reach for docker-expert when Dockerfiles balloon past acceptable limits, containers run as root, Compose services miswire networks or volumes, or production deployment patterns need correction. The skill applies current industry practices to optimize build contexts, layer caching, health checks, and orchestration YAML instead of generic container advice. Use it during Dockerfile rewrites, incident response on image vulnerabilities, and Compose stack design for APIs, workers, and databases heading to staging or production.828installs58Draw Iodraw-io is a Claude Code skill from davila7/claude-code-templates for AI-assisted creation and maintenance of draw.io diagrams. The skill works with `.drawio` XML files, enforces consistent styling and layout, adjusts element positions programmatically, and automates PNG conversion with transparent backgrounds for presentations and docs. It targets architecture diagrams, flowcharts, and technical documentation visuals, including font-family consistency for Quarto slide decks. Developers reach for draw-io when they need editable diagram source files plus export-ready PNGs without manual draw.io GUI work.821installs59Pdfpdf is a Claude Code skill from davila7/claude-code-templates that equips coding agents to open PDF files, summarize long documents, and pull structured fields such as tables, headings, and key facts into the active development session. The skill is designed for moments when a PDF blocks progress—API PDFs, compliance packets, exported architecture diagrams, or vendor datasheets that never lived in the repository. Instead of switching to a separate viewer or OCR pipeline, the agent ingests the file in-context and returns quotable excerpts developers can paste into tickets, tests, or design notes. pdf fits backend and full-stack work where requirements arrive as attachments rather than markdown. Reach for pdf when a binary document is the source of truth and you need fast, accurate extraction before writing code or updating docs.813installs60Security Compliancesecurity-compliance is a davila7/claude-code-templates skill that provides an Incident Response Report Template for documenting security and production incidents. The template includes incident ID, P0–P3 severity, status lifecycle (Open, Contained, Resolved, Closed), incident commander assignment, UTC timestamps, executive summary, and classification fields for incident type, attack vector, and affected assets. Developers and responders reach for security-compliance immediately after malware, phishing, data breach, DDoS, or unauthorized access events when a standardized report must be produced fast. The skill accelerates post-incident documentation so teams capture classification, impact, and resolution details in a consistent compliance-ready format.812installs61Senior Ml Engineersenior-ml-engineer is an LLM integration guide skill from davila7/claude-code-templates that encodes senior ML/AI engineer practices for production systems. The skill emphasizes designing for 10x scalability headroom, 99.9% uptime reliability, full observability, input validation, encryption, access control, and audit logging from day one. Advanced patterns cover distributed processing, strategic caching, batch processing, and enterprise-scale fault tolerance for AI backends. Developers reach for senior-ml-engineer when building LLM-powered APIs or agents and need architecture decisions that survive production load rather than prototype-quality integration code.799installs62Excalidrawexcalidraw is a claude-code-templates skill enforcing subagent delegation for *.excalidraw and *.excalidraw.json files. Main agents must not read diagram JSON directly because single files range from 4k to 22k tokens and the largest can exceed read limits; multiple diagrams quickly exhaust context. Triggers include diagram, flowchart, and architecture visualization requests. Subagents isolate token-heavy JSON manipulation while the parent agent keeps concise summaries. Developers reach for excalidraw when agents must create or edit Excalidraw architecture diagrams without blowing the context window.794installs63Marketing Ideasmarketing-ideas is a davila7 Claude Code templates skill acting as a marketing strategist with a library of 140 proven marketing ideas for SaaS and software products. When triggered by phrases like marketing ideas, growth ideas, or ways to promote, the skill asks about product, audience, and resources then recommends targeted tactics from its categorized playbook rather than generic advice. Developers reach for marketing-ideas when they have a shippable product but need structured inspiration for acquisition channels, content plays, or low-budget promotion experiments. The workflow is consultative: clarify context, filter the 140-idea library to fit constraints, and output actionable strategies developers can execute or hand to a growth teammate without building a campaign from a blank page.767installs64Statistical Analysisstatistical-analysis is a Claude Code skill from davila7/claude-code-templates that encodes systematic guidance for validating statistical assumptions across common tests before interpreting results. The readme centers on principles such as always check assumptions before interpreting test results, use multiple diagnostic methods including visual and formal tests, and document violations with remedial actions in analysis reports. Coverage spans independence of observations and other cross-test assumptions with explicit diagnostic workflows rather than one-line reminders. Developers reach for statistical-analysis when an agent or notebook proposes regressions, t-tests, or similar inference and the team needs a structured assumption audit before trusting coefficients, p-values, or automated summaries. The skill supports robustness reasoning when violations occur under known conditions. It is reference-grade statistical QA for analysis pipelines, not a statistics textbook or automated test runner.764installs65Devops Iac Engineerdevops-iac-engineer is a DevOps & CI/CD agent skill from davila7/claude-code-templates that helps teams design, implement, and maintain cloud infrastructure using Infrastructure as Code principles. The skill covers Terraform modules, Kubernetes deployments, scalable cloud architectures, CI/CD pipeline design, observability solutions, and security-first DevOps practices with site reliability engineering guidance. Developers reach for devops-iac-engineer when building cloud architectures, deploying containerized applications, setting up CI/CD automation, or implementing monitoring and DevSecOps controls. Reference navigation points to dedicated terraform.md guidance for IaC patterns. The skill assumes skilled engineers who need structured cloud-native recommendations across provisioning, deployment, and operational reliability rather than single-file application logic.755installs66CopywritingCopywriting is a claude-code-templates agent skill that guides coding agents to write clear, compelling marketing copy for product pages. It covers homepages, landing pages, pricing pages, feature pages, about pages, and product pages—gathering page purpose, audience, and value props before drafting headlines, body copy, and CTAs. Triggers include requests to write copy, rewrite a page, improve marketing text, or refine headline and CTA language. Developers reach for Copywriting when shipping SaaS or product sites and needing conversion-oriented page text; the skill routes email copy to email-sequence and popup copy to popup-cro siblings.753installs67Browser Automationbrowser-automation in davila7/claude-code-templates documents battle-tested browser automation for testing, scraping, and AI agent control, sourced from vibeship-spawner-skills under Apache 2.0. The skill recommends Playwright as the default framework while covering Puppeteer when stealth or Chrome-only constraints apply. It emphasizes selectors, explicit waiting strategies, and anti-detection patterns that separate flaky scripts from production-grade automation. Developers reach for browser-automation when wiring E2E tests, building scrapers, or giving agents dependable browser tools instead of one-off scripts that break on timing or layout changes.738installs68Context7 Auto Researchcontext7-auto-research is a Claude Code skill that automatically retrieves up-to-date official documentation for libraries and frameworks through the Context7 API and injects it into the active conversation. Install globally with `npx skills add -g BenedictKing/context7-auto-research`, optionally configure an API key, then invoke naturally when asking about React, Next.js, Prisma, or other packages. Developers reach for context7-auto-research when stale training data risks wrong imports, deprecated APIs, or missing config flags during feature work. The skill keeps answers anchored to published docs instead of guesswork from memory.721installs69Agent Memory Systemsagent-memory-systems is a davila7/claude-code-templates skill sourced from vibeship-spawner-skills (Apache 2.0) for architecting agent memory beyond the context window. It covers short-term context-window memory, long-term vector stores, and cognitive patterns that organize storage around retrieval—not just persistence—emphasizing chunking, embedding, and retrieval strategies. Developers reach for agent-memory-systems when agents forget prior sessions, recall irrelevant facts, or need typed long-term memory before production deployment. The skill frames memory as a retrieval problem where stored facts only matter when the right chunk surfaces at the right time.715installs70Pubmed Databasepubmed-database is a Claude Code skill providing a reference for the NCBI E-utilities REST API at eutils.ncbi.nlm.nih.gov/entrez/eutils/. The skill covers esearch.fcgi and related endpoints, API key registration for raised rate limits, and request parameter patterns for PubMed queries. Without an API key, NCBI enforces 3 requests per second; with a registered key, the limit increases to 10 requests per second. Developers reach for pubmed-database when building health-tech agents, research pipelines, or citation tools that need direct PubMed access from Claude Code, Cursor, or custom agents.713installs71Content Creatorcontent-creator is a content planning skill from davila7/claude-code-templates that generates structured monthly content calendars for consistent publishing across blog, LinkedIn, Instagram, email newsletter, and video channels. Each week includes dated entries with platform, topic, keywords, captions, hashtags, CTAs, owner assignments, and publish-status checkboxes. Developers and technical marketers use content-creator when they need a repeatable editorial template with traffic, lead, and engagement goals rather than ad-hoc posting. The template opens with monthly goals and key campaign fields, then breaks work into daily multi-platform schedules.709installs72Frontend Dev Guidelinesfrontend-dev-guidelines is a skill from davila7/claude-code-templates that gives instant access to proven React, TypeScript, MUI, and React Hook Form patterns for common UI work. The readme documents authentication with useAuth, DataGrid setups, dialogs, and form flows with explicit guidance such as never making direct API calls for auth. Developers reach for it when scaffolding SaaS dashboards, admin panels, or internal tools that repeat the same MUI and hook-form structures. The skill encodes copy-paste-ready conventions instead of one-off component reinvention.702installs73Pdf Processing Propdf processing pro is a Claude Code skill from davila7/claude-code-templates offering a production-oriented PDF processing toolkit with bundled scripts, validation, and error handling. The skill uses pdfplumber and included analysis scripts to extract text, parse forms, handle tables, run OCR workflows, and process PDF batches at volume. Developers reach for pdf processing pro when building document ingestion pipelines, invoice parsers, or compliance workflows that must survive messy real-world PDFs without silent failures. Quick-start examples cover single-page text extraction and form analysis via command-line scripts with explicit validation steps.699installs74Perplexity Searchperplexity-search is a davila7/claude-code-templates skill that configures Perplexity models via OpenRouter for agent web research. It provides a copy-ready .env template with OPENROUTER_API_KEY, optional DEFAULT_MODEL choices including sonar-pro, sonar-pro-search, sonar, sonar-reasoning-pro, and sonar-reasoning, plus DEFAULT_MAX_TOKENS and DEFAULT_TEMPERATURE settings. Developers reach for perplexity-search when grounding agent answers with live web retrieval without building a custom Perplexity client. The bundled model comparison guide explains when to pick each sonar variant for search-heavy versus reasoning workloads.675installs75Claude Code Guideclaude code guide is a comprehensive reference skill from davila7/claude-code-templates for configuring and using Claude Code effectively. The skill synthesizes best practices for creating CLAUDE.md project guideline files with commands, coding standards, and architecture notes that steer the agent on every session. It documents thinking-keyword prompting strategies that trigger deeper reasoning, plus debugging techniques for diagnosing stuck or misbehaving agent sessions. Developers reach for claude code guide when onboarding to Claude Code, setting up a new repository for agent-assisted development, or troubleshooting sessions that ignore project conventions or produce shallow responses.673installs76Ai Agents Architectai-agents-architect is a Claude Code skill sourced from vibeship-spawner-skills (Apache 2.0) that acts as an AI Agent Systems Architect. It designs autonomous agents with tool use, memory systems, planning strategies, and multi-agent orchestration while planning for graceful degradation and clear failure modes. Developers reach for ai-agents-architect when building agents that must balance autonomy with oversight—knowing when to ask for help versus proceed independently. The skill triggers on requests to build agents, design tool-calling flows, add memory, or coordinate multiple agents. It emphasizes architecture before implementation so agents remain controllable under real-world failure conditions.670installs77Api Integration Specialistapi integration specialist is a davila7/claude-code-templates skill providing expert guidance for integrating external APIs with production-ready patterns. The skill covers OAuth flows, API key management, request and response transformation, retry logic, rate limiting, and robust error handling for REST endpoints, GraphQL APIs, and webhooks—including services like Stripe, Twilio, and SendGrid. Developers reach for api integration specialist when building API client libraries, wiring payment or messaging providers, or hardening third-party service calls beyond basic fetch wrappers.661installs78Clinical Reportsclinical-reports is a Claude Code skill from davila7/claude-code-templates that walks an agent through a full clinical case report skeleton—title, author block, MeSH keywords, and abstract subsections for introduction, patient concerns, and diagnosis. The template mirrors standard medical publishing conventions so output can move toward journal submission with minimal reformatting. Developers in health tech, bioinformatics, or clinical research reach for clinical-reports when they must turn patient encounter notes or study observations into a consistent case-study document rather than free-form markdown. The skill does not perform diagnosis; it structures narrative and metadata fields the author supplies.655installs79Nestjs Expertnestjs-expert is a Nest.js framework expert skill for Claude Code that applies enterprise Node.js patterns when designing modules, dependency injection graphs, middleware, guards, interceptors, and authentication. The skill covers Jest and Supertest testing strategies plus TypeORM, Mongoose, and Passport.js integration when wiring persistence and auth. Developers reach for nestjs-expert when debugging complex DI failures, choosing architecture for new features, or hardening tests and performance in an existing Nest application. The skill proactively recommends switching to a more specialized expert when another framework fit is clearer.651installs80Pytorch Lightningpytorch-lightning is a davila7 claude-code-templates skill focused on PyTorch Lightning callbacks that capture non-essential training logic outside the LightningModule. It documents built-ins such as ModelCheckpoint with save_top_k, monitor, and save_last options, plus EarlyStopping, learning rate monitoring, and logging integrations wired through L.Trainer. Developers reach for pytorch-lightning when they need reliable checkpoint directories, validation-driven model selection, or hook-based logging without cluttering forward and training_step code. Examples use lightning.pytorch.callbacks imports and Trainer callback lists ready to paste into training scripts.649installs81Frontend Patternsfrontend-patterns is a version 1.0 frontend development skill from claude-code-templates for developers using React and Next.js with AI agents. The skill encodes modern UI guidance such as composition over inheritance with typed Card/CardHeader components, state management conventions, performance optimization tactics, and interface best practices illustrated with TypeScript examples. Developers reach for frontend-patterns when agent-generated UI code needs consistent component structure, variant props, and performance-conscious patterns instead of one-off JSX. The patterns emphasize composable children, explicit className strategies, and Next.js-aware frontend architecture suitable for production SaaS, extension, and mobile web surfaces.629installs82Agent Evaluationagent-evaluation is a Claude Code skill from davila7/claude-code-templates, sourced from vibeship-spawner-skills under Apache 2.0, for testing and benchmarking LLM agents where even top agents achieve less than 50% on real-world benchmarks. The skill treats agent evaluation as fundamentally different from traditional software testing because identical inputs can yield different outputs and correctness often has no single answer. It covers behavioral testing, capability assessment, reliability metrics, and production monitoring workflows. Developers reach for agent-evaluation when shipping coding agents or autonomous workflows and need structured evaluation before trusting benchmark scores alone.620installs83Nextjs Best PracticesNext.js Best Practices is an agent skill that embeds official App Router principles directly into your Claude, Cursor or Codex workflow. It provides decision trees, pattern tables, and concrete rules for when to use Server Components versus Client Components, which data fetching strategy to apply, and how to structure routes and layouts. The skill gives developers immediate, context-aware guidance on caching behavior, component splitting, file conventions, and data flow so that every file created follows current Next.js recommendations without needing to leave the editor or consult external docs. It is especially valuable for solo builders who want to ship production-grade Next.js applications without accumulating anti-patterns around rendering, performance, or routing organization.595installs84Peer ReviewPeer-review is an agent skill that applies a standardized checklist of common methodological and statistical pitfalls to any research-oriented output. It scans for p-value misuse, inappropriate test selection, missing effect sizes, data dredging, poor visualization practices, and other frequent errors that undermine credibility. The skill returns specific, constructive feedback with recommended fixes such as reporting confidence intervals, applying multiple-testing corrections, and using equivalence tests. Ideal for solo builders producing data-driven features, academic-style documentation, or scientific manuscripts who want to eliminate sloppy statistical claims before they reach users, collaborators, or publication.584installs85Game Developmentgame-development is an agent skill that supplies a collection of Claude-optimized starter templates and implementation patterns for building 2D games. It gives solo developers a fast on-ramp to create canvas-based games by providing consistent structure for game loops, asset handling, state management, and input systems. Builders use it when they want to skip boilerplate and focus on unique gameplay instead of fighting with basic engine code. The templates are written to be easily understood and extended by large language model coding agents, reducing token waste and context-switching. This skill is especially useful for indie creators shipping small web games, jam entries, or experimental interactive experiences without a full game engine.583installs86Prompt Engineeringprompt-engineering is an expert guide to LLM prompt patterns bundled in davila7/claude-code-templates. It covers few-shot learning with 2–5 input-output examples, reasoning format control, edge-case handling, and token budgeting tradeoffs when teaching models by demonstration instead of prose rules. Developers reach for prompt-engineering when agent behavior is inconsistent, formatting drifts, or prompts need debugging across Claude Code, Cursor, or generic LLM workflows. The skill emphasizes reliability and controllability for coding agents rather than one-off chat tips. Use it while designing system prompts, tool instructions, or reusable agent templates that must survive production variability.583installs87Latex Posterslatex-posters is an agent skill that supplies a ready-to-use baposter LaTeX template for creating large-format academic and research posters. It includes pre-configured document class settings, essential packages for graphics, mathematics, tables, QR codes and hyperlinks, plus sensible defaults for spacing, colors and grid layout. Builders simply fill in their content and compile to PDF. The template eliminates manual layout debugging so you can focus on communicating results clearly. Ideal when you need to produce conference posters, demo summaries, or visual project overviews quickly and consistently.580installs88Cto AdvisorCTO Advisor is an agent skill that generates and manages Architecture Decision Records (ADRs). It provides a structured template and workflow for documenting significant technical choices including the problem context, decision drivers, options considered, chosen solution with justification, positive and negative consequences, and links to related artifacts. Solo builders and small teams use it to preserve decision rationale that would otherwise be lost in chat histories or tribal knowledge. This prevents repeated debates, reduces onboarding time, and gives future-you or collaborators clear reasoning when revisiting architecture months later. The skill outputs ready-to-commit markdown files that become living documentation alongside your codebase.577installs89Gdpr Dsgvo Expertgdpr-dsgvo-expert is a senior-level agent skill for solo and indie builders who process personal data in the EU or Germany and need structured GDPR and DSGVO guidance instead of vague “be compliant” chat advice. It is invoked for compliance assessments, privacy audits, data-protection planning, and verification against Articles covering lawful basis, special categories, consent, information duties, access, rectification, erasure, restriction, portability, and objection. The skill encodes framework-style checklists and auditor thinking so your agent can draft gap analyses, DPIA outlines, and governance recommendations aligned with accountability obligations. It does not replace licensed legal counsel for binding interpretations, but it compresses expert audit methodology into repeatable agent sessions during Validate scoping and Ship security hardening.577installs90Web Design GuidelinesWeb Design Guidelines is an agent skill that reviews your frontend UI code against the latest Vercel Web Interface Guidelines. It automatically fetches the current rule set from the official source, reads the files or patterns you specify, checks every rule for compliance, and returns concise findings. Builders use it to keep interfaces consistent, accessible, and high-quality without manually tracking dozens of evolving best practices. The skill eliminates guesswork when reviewers or users ask you to validate design, accessibility, or overall UX. It works on any web codebase and integrates directly into your existing agent workflow.576installs91Product Manager ToolkitProduct Manager Toolkit is a reusable Claude Code template that produces professional-grade Product Requirements Documents. It guides developers and AI agents through a systematic process covering executive summary, deep customer problem analysis using the 5W framework, market sizing, business case development, solution mapping, scoping decisions, and success metrics. The structured format eliminates ambiguity that typically causes AI agents to build the wrong features or waste cycles on out-of-scope work. Developers copy the template into their workspace and invoke it with project context to receive a complete, stakeholder-ready PRD that serves as a reliable contract between the developer and their coding agents.567installs92Agent Manager SkillAgent Manager Skill is a Python CLI tool that lets you run, assign work to, and supervise multiple local agents simultaneously. Each agent runs inside its own tmux session so they operate independently without interfering. You can start or stop agents, tail their logs in real time, run health checks, and hand them concrete tasks using simple commands. The skill is especially useful when you need parallel agent execution for research, maintenance, content generation, or monitoring workflows. It stores configuration under an agents/ directory and works with any CLI-based agent that can be launched from a script. Once installed via git clone, you gain a repeatable way to scale from one agent to many without losing visibility or control.563installs93Research Lookupresearch-lookup is a Claude Code skill from davila7/claude-code-templates that wraps a Python ResearchLookup client for scientific and technical literature queries. The skill automatically selects between a fast Sonar Pro Search model for simple lookups and deeper reasoning models for complex questions, with optional manual override and batch query processing. Developers reach for research-lookup when writing technical docs, evaluating algorithms, or grounding agent responses in published research without leaving the coding environment. The bundled example script demonstrates automatic model selection, manual overrides, and integration with scientific writing workflows.562installs94Product StrategistProduct Strategist is an agent skill that creates aligned OKRs by cascading high-level company strategy into specific, measurable objectives and key results across growth, retention, and product focus areas. developers and developers use it to turn vague product direction into concrete, trackable goals with built-in templates that include user acquisition, churn reduction, NPS improvement, and activation metrics. It replaces ad-hoc goal setting with a repeatable generation process that fills in current and target values, produces time-bound outcomes, and delivers structured JSON that can be directly imported into roadmaps or handed off to coding agents. This keeps operators focused on execution instead of spending hours inventing metrics that actually ladder up to business success.561installs95SeabornSeaborn is a Python agent skill that teaches practical, copy-paste visualization workflows on top of pandas, matplotlib, and numpy. developers use it when they need credible charts during exploration, feature engineering, or stakeholder-facing reports without relearning seaborn’s API surface each time. The readme walks through exploratory data analysis (pairplots with hue and corner layouts, faceted kde displots), correlation matrices with upper-triangle masks, and scientific multi-panel figure setups. It emphasizes sensible defaults—hue by target or condition, col_wrap faceting, common_norm control, and export discipline—so agents produce consistent filenames and DPI rather than inline-only plots. Install it when your agent is writing analysis notebooks, validating datasets before modeling, or embedding visualization snippets into backend jobs that emit PNGs. It is procedural knowledge, not a running service: you still need your CSV or DataFrame in the repo and matplotlib configured in the environment.560installs96Feature Design Assistantfeature-design-assistant is a pre-implementation planning skill from davila7/claude-code-templates that guides structured feature design through natural dialogue. Phase 1 explores the codebase for project structure, tech stack, existing patterns, related modules, and recent changes in the target area. The skill gathers requirements collaboratively, validates assumptions, and produces a fully formed design and spec before agents rewrite large sections. Use it when planning new features, designing architecture, or making significant codebase changes where jumping straight to code risks mismatched conventions or missed dependencies.559installs97Agent Memory McpAgent Memory MCP is an integration skill that boots a searchable, persistent memory server for AI coding agents. developers shipping agent-heavy products use it in Build to stop re-explaining architecture every session. Setup clones the agentMemory repo into the skills workspace, installs Node dependencies, compiles, and starts the MCP server scoped to a project id and absolute workspace path. Exposed tools let the agent query prior patterns (“authentication”) and record new decisions with structured types and tags. The model fits multi-week builds and later Operate work where prior decisions must stay discoverable. It requires local Node tooling and comfort running MCP alongside your editor. It complements ad-hoc markdown notes by making memory machine-retrievable through standard MCP calls rather than brittle copy-paste context.558installs98Tailwind Patternstailwind-patterns is a Tailwind CSS v4 reference skill for developers migrating from v3 or starting fresh with CSS-native configuration. It documents the shift from tailwind.config.js to @theme directives, PostCSS plugin to the Oxide engine advertised as 10x faster, always-on JIT, and CSS-native features replacing the old plugin system. Developers reach for this skill when building responsive layouts with container queries, architecting design tokens, or avoiding discouraged @apply-heavy patterns. Allowed tools include Read, Write, Edit, Glob, and Grep for applying patterns directly in project CSS files.554installs99I18n LocalizationThe i18n-localization skill packages a small Python scanner that audits your frontend and backend source for strings that should live in translation catalogs instead of hardcoded literals. developers shipping SaaS or APIs with React, Vue, or Python surfaces invoke it when adding locales or cleaning up tech debt before release. It contrasts suspicious capitalization-heavy literals against established i18n call patterns so your agent can prioritize refactors. It is a focused checker—not a full TMS integration—meant to complement react-i18next, vue-i18n, or gettext workflows. Run it after feature work or before merge when multilingual support is on the roadmap, and feed findings into your normal PR workflow.550installs100Context Window Managementcontext-window-management is a context engineering skill sourced from vibeship-spawner-skills that teaches strategies for summarization, trimming, routing, and avoiding context rot in LLM applications and long agent sessions. Developers reach for it when conversations hit token limits, lose critical information mid-dialogue, or degrade as more tokens accumulate without better results. The skill treats context as a finite resource with diminishing returns and guides curation of the right information density rather than maximizing raw token count. Triggers include context window, token limit, context management, context engineering, and long context keywords.548installs101Seo Fundamentalsseo-fundamentals is a repository SEO audit skill from davila7/claude-code-templates backed by a Python seo_checker.py script. The checker scans HTML files and JSX/TSX React page components that represent public routes, verifying meta titles and descriptions, Open Graph tags for social sharing, heading hierarchy, and image alt attributes. Run it with `python seo_checker.py <project_path>` before shipping or refreshing marketing pages. Use seo-fundamentals when you need a repeatable on-repo pass over public page files instead of manual tag inspection or external-only Lighthouse runs.546installs102ArchitectureContext Discovery is an agent skill that systematically gathers project context through a structured question hierarchy before any architecture is proposed. It uses a five-category questionnaire covering scale, team, timeline, domain, and constraints, then maps responses to a Project Classification Matrix that recommends appropriate architectures for MVP, SaaS, or Enterprise projects. The skill helps developers avoid over-engineering by matching architecture complexity to actual needs, providing concrete examples such as Next.js API routes for simple MVPs or NestJS for mid-scale SaaS applications. It ensures architecture decisions are grounded in real project parameters rather than assumptions.539installs103HumanizerHumanizer is an agent skill that removes predictable patterns typical of large language models from any text you give it. It rewrites content to sound more natural, varied, and human by targeting 24 specific signs of AI writing identified by Wikipedia's WikiProject AI Cleanup. Developers use it on marketing copy, documentation, blog posts, landing pages, and any prose that will face public scrutiny. The skill is based on the insight that LLMs favor statistically common phrasing that applies broadly, resulting in telltale generic language. After running Humanizer, your text loses robotic repetition, inflated significance, hedging, and other machine-like artifacts while preserving original meaning and improving readability. It installs directly into Claude Code's skills directory and can be invoked with a simple slash command or by asking Claude to humanize text.535installs104Brainstormingbrainstorming is an agent skill from the davila7 Claude Code templates that forces a design conversation before your agent writes code. It treats implementation as the last step: first align on project context, then refine the idea through focused questions—one per turn—about purpose, constraints, and success criteria. When the problem space is clear, it explores two or three architectural approaches with explicit trade-offs and a stated recommendation, then drafts the design in digestible sections so you can correct course early. developers benefit because it prevents expensive rework on features, UI components, integrations, and behavioral tweaks across any stack. Use it whenever the spec lives only in chat, when multiple valid designs exist, or when you want the agent to behave like a patient product partner instead of jumping straight to patches.532installs105PolarsPolars is an agent skill that teaches developers how to write efficient Polars code for real datasets: prefer lazy evaluation, push filters and projections early, and avoid Python functions that break vectorization and parallel plans. It is for anyone shipping Python backends, CLI data tools, or agent workflows that read CSV/Parquet, join tables, and aggregate metrics without drowning in memory or wall time. Use it when you or your coding agent is drafting transforms, reviewing a slow pipeline, or refactoring pandas-style eager scripts into production-grade Polars. The guide walks through concrete bad-vs-good patterns—eager reads versus scan plus collect, late filters versus early select/filter, and expression-native logic versus per-row Python callbacks—so outcomes are faster IO, smaller intermediate frames, and query plans the engine can optimize.532installs106Nodejs Best Practicesnodejs-best-practices is a principles-first skill from davila7/claude-code-templates for Node.js development in 2025. Rather than dumping fixed templates, it walks developers through framework selection—Hono for edge and serverless, Fastify for high-performance APIs, Express for enterprise teams—and async error-handling, security defaults, testing hooks, and module layout based on project context. The skill explicitly asks for user preferences when requirements are unclear and avoids one-size-fits-all defaults. Developers reach for it when standing up or refactoring Node services where framework tradeoffs, production readiness, and maintainable structure matter. Allowed tools include Read, Write, Edit, Glob, and Grep for codebase-aware guidance.531installs107Audiocraft Audio Generationaudiocraft-audio-generation is a davila7/claude-code-templates guide for fine-tuning Meta AudioCraft MusicGen on custom audio corpora. The documented prepare_dataset workflow uses torchaudio to emit an output_dir with audio/ numbered WAV files such as 0001.wav and a metadata.json mapping paths to descriptions. Developers reach for audiocraft-audio-generation when adding bespoke generative music to apps or demos and need reproducible 32 kHz mono dataset prep before training. Use it during model customization rather than only calling a hosted music API with no local fine-tune pipeline.529installs108PineconePinecone is an agent skill that teaches developers how to deploy and operate Pinecone vector indexes in production using the official Python client. It walks through creating serverless indexes with ServerlessSpec across AWS, GCP, or Azure regions—ideal when traffic is spiky and you want auto-scaling without managing pods. It also covers pod-based PodSpec setups with pod types, replica counts, and dedicated capacity when you need stable latency and higher query throughput. The guide adds hybrid search by upserting dense embeddings alongside sparse token indices so retrieval can blend semantic similarity with lexical signals. Use it while wiring RAG, agent memory, or product search in build and integrations, then revisit at ship and operate when you promote an index from dev keys to real traffic, regions, and cost controls.529installs109Database Designdatabase-design is a Claude Code skill from davila7/claude-code-templates that turns natural language product requirements into production-oriented database artifacts. The skill produces entity-relationship structures, foreign-key relationships, index recommendations, and step-by-step migration plans suitable for PostgreSQL, MySQL, and similar relational stacks. Developers reach for database-design when bootstrapping a new service, refactoring a monolith schema, or translating PRD tables and workflows into normalized tables before writing ORM models or raw SQL migrations. With 508 installs on skills.sh, it fits early backend sprints where schema mistakes are expensive to unwind later.527installs110Seo AuditSEO Audit is an agent skill that turns vague ranking complaints into a structured diagnostic report. It first gathers context about your site type, business goals, target keywords, current traffic, and known issues. Then it applies a fixed 5-step framework starting with crawlability and indexation, moving through technical foundations, on-page optimization, content quality, and finally authority and links. The skill surfaces specific, prioritized problems such as robots.txt blocks, missing meta tags, slow Core Web Vitals, thin content, or missing schema so you can fix the highest-impact items first. Developers use it when they hear "why am I not ranking" or want a health check before a launch or content push. Because it follows a repeatable expert framework instead of ad-hoc prompting, it consistently catches issues that generic chat requests miss.525installs111Agent Tool Builderagent-tool-developer is an agent skill for developers shipping AI products who need tools that work on the first call—not loops, silent failures, or mystery 10x token bills. It treats the schema and natural-language description as the real API surface, because models never execute your Python; they only choose arguments against what you documented. You get patterns for crisp JSON Schema, examples that steer usage, validation boundaries, and errors that teach the model how to retry. MCP is framed as the practical lingua franca when you want the same tool usable across Claude Code, Cursor, and other hosts. Use it while defining new capabilities in Build, and again in Ship or Operate when tightening production tool contracts after real agent traces.522installs112Docxdocx is an agent skill published in the davila7/claude-code-templates collection on skills.sh with 512 recorded installs and catalog rank 9386. It lets Claude Code agents produce clean, structured Microsoft Word .docx files directly from conversation context or explicit generation instructions. Developers reach for docx when they need specifications, reports, proposals, or handoff documents in Word format rather than plain Markdown or PDF. The skill fits documentation and office-document workflows inside agent sessions where formatted headings, sections, and Word-native structure matter to stakeholders. Because output is a standard .docx file, recipients can edit in Microsoft Word, Google Docs, or compatible suites without conversion pipelines.521installs113Marketing Demand AcquisitionMarketing Demand Acquisition is a lightweight agent skill centered on a Python CAC (Customer Acquisition Cost) calculator for founders validating whether marketing spend is efficient. Install it when you have campaign spend and customer counts by channel—or only totals—and need blended and channel-specific CAC without spinning up a full analytics stack. The workflow accepts structured channel dictionaries, computes per-channel and blended figures, and formats output for growth reviews, pricing sanity checks, or board-style summaries. It suits SaaS, content products, and small e-commerce shops comparing ads, SEO-assisted signups, or partnerships in one view. It does not replace attribution tooling or cohort LTV analysis; it gives fast, explicit math your agent can re-run as you update monthly numbers.521installs114Skill Creation Guideskill-creation-guide is a Claude Code skill from davila7/claude-code-templates that walks developers through generating agent skills with consistent structure, metadata, and quality checks matching Skillselion conventions. The skill encodes naming, frontmatter, trigger phrases, and workflow sections so new skills install cleanly into Claude Code and surface correctly in skill marketplaces. With 513 installs on skills.sh, it targets developers who already ship agent workflows and want a repeatable template instead of hand-writing every SKILL.md. Reach for skill-creation-guide when bootstrapping a new skill repo, standardizing an existing skill library, or onboarding teammates to the same agent-skill format.521installs115Modal Serverless Gpumodal-serverless-gpu is an agent skill that lets Claude Code, Cursor, and similar agents instantly provision and use GPU compute on Modal's serverless platform. Instead of managing always-on instances or wrestling with cloud provider consoles, you describe the workload and the skill spins up the right GPU, runs your code, then tears everything down. It is built for developers and small teams who need occasional heavy AI compute without monthly bills or DevOps overhead. The skill wraps Modal's Python SDK into reusable agent commands so you stay in your editor while the infrastructure appears only when needed and disappears when the job finishes.518installs116Notion Knowledge CaptureNotion Knowledge Capture is an agent skill for developers who lose decisions in chat threads and need a repeatable path into Notion. It walks you from clarifying audience and capture type through picking the right database template, pulling existing context with Notion search and fetch, drafting structured pages with summaries, source links, tags, and owners, then linking hubs and updating status as sources evolve. The workflow assumes Notion MCP connectivity and documents a pause-and-setup sequence—adding the remote MCP, enabling rmcp_client, and completing OAuth—when calls fail. Multi-phase by nature: you capture during Build docs sprints, Grow content operations, and Operate iteration when incidents spawn new runbooks. Best for agent-forward teams already on Notion; skip if you refuse cloud knowledge bases or cannot enable MCP OAuth in your Codex environment.518installs117Obsidian MarkdownTeaches Claude to write and refactor Obsidian-flavored markdown with wikilinks, callouts, tags, YAML front matter, and vault-safe structure so content, CLI, and agent knowledge bases stay linked, searchable, and easy to maintain during build.515installs118Content Research Writercontent-research-writer is an agent skill that conducts comprehensive online research on any topic and then composes well-structured, authoritative long-form content. developers and teams use it to create blog posts, landing-page copy, newsletters, and SEO articles that feel human-written yet contain accurate, up-to-date information. The skill first gathers and synthesizes sources, then applies tone, structure, and citation rules before delivering the final draft. Because it bundles research and writing, it prevents hallucinated facts and generic AI slop while dramatically shortening the content creation cycle. Install it once and invoke it whenever you need trustworthy marketing or educational content without becoming a full-time researcher.512installs119React Ui Patternsreact-ui-patterns is a collection of reusable React component templates and UI patterns that developers and developers use to speed up frontend work. It gives you battle-tested starting points for common interface elements so you spend less time on boilerplate and more time on product logic. The skill works seamlessly with Claude Code, Cursor, and other agentic coding tools by providing clear, well-structured examples that agents can easily understand and extend. Whether you are building a SaaS dashboard, a marketing site, or an AI tool interface, these patterns help maintain consistency and quality without forcing you to reinvent common UI solutions every time.507installs120Create Plancreate-plan from davila7/claude-code-templates turns a user prompt into a single actionable implementation plan delivered in the final assistant message. The skill operates entirely in read-only mode without writing or updating files during planning. It scans README.md, docs/, CONTRIBUTING.md, and ARCHITECTURE.md, skims files most likely to be touched, and identifies constraints including language, frameworks, CI test commands, and deployment shape before asking follow-up questions only when necessary. Developers reach for create-plan when they explicitly ask for a plan related to a coding task and want a concise, ordered execution roadmap their agent can follow.503installs121Research Grantsresearch-grants from davila7/claude-code-templates helps developers quickly surface relevant grant opportunities, eligibility criteria, and application requirements using Claude. Listed on skills.sh with 503 installs, the skill supports early funding research for technical projects and research proposals without requiring manual grant-database navigation. Developers reach for research-grants when exploring whether their project qualifies for research grants, startup funding programs, or institutional awards and need a fast summary of requirements and deadlines before drafting applications.503installs122Screenshot Feature Extractorscreenshot-feature-extractor is a Claude Code skill from davila7/claude-code-templates (rank 29 on skills.sh, 480 installs) that turns UI screenshots into structured feature lists and implementation prompts. Developers reach for screenshot-feature-extractor when they have a design reference image—mockup, competitor screen, or wireframe—and need agent-ready specs instead of manually listing components. The skill bridges design references and coding agents by producing actionable feature breakdowns Claude or Cursor can implement. With 480 installs, it is a common validate-phase shortcut for teams prototyping UI from visual inputs before writing frontend code.502installs123Planning With Filesplanning-with-files is an agent skill that turns high-level ideas into persistent, file-based plans stored directly in your repository. Instead of losing context in chat threads, it writes clear, structured markdown files that Claude, Cursor, or other agents can read and follow. This keeps instructions accurate across long sessions and multiple files. Developers use it to maintain living specifications, task breakdowns, and decision records that stay synchronized with the codebase. The skill emphasizes creating concrete deliverables that remove ambiguity before coding starts, making it especially valuable for complex features or when wearing every hat as a founder.501installs124Rag Implementationrag-implementation is a Claude Code skill from davila7/claude-code-templates, sourced from vibeship-spawner-skills under Apache 2.0, that guides developers through production-grade retrieval-augmented generation. The skill covers document chunking strategies, embedding selection, vector store configuration, retrieval policies, reranking, and grounded answer flows that keep LLM responses tied to source material. It explicitly addresses failure modes of naive vector-search-only approaches and teaches when RAG helps versus when simpler context injection suffices. Developers reach for rag-implementation when building Q&A bots, internal knowledge assistants, or document search features that must scale beyond prototype quality. The specialist persona reflects experience serving high query volumes over large document sets.499installs125React Patternsreact-patterns is a frontend skill from davila7/claude-code-templates covering modern React architecture for production applications. The skill defines server, client, presentational, and container component types with clear state boundaries, emphasizing one responsibility per component, props down and events up, and composition over inheritance. It covers hooks usage, performance optimization, and TypeScript best practices with Read, Write, Edit, Glob, and Grep tool access for agent sessions. Developers reach for react-patterns when Claude Code or Cursor sessions need consistent guidance on component design rather than ad hoc React advice. The skill suits teams standardizing how agents generate and refactor React code across a codebase.498installs126Executing Marketing Campaignsexecuting-marketing-campaigns is a davila7/claude-code-templates skill that guides an AI agent through planning and executing a coordinated marketing campaign around a product launch or feature release. Listed on skills.sh with 477 installs and catalog rank 30, it structures channel selection, messaging, sequencing, and rollout tasks so developers do not improvise launch communications file by file. The skill fits release weeks when a feature ships and teams need announcement posts, email copy, social schedules, and campaign checklists produced in one workflow. Developers reach for executing-marketing-campaigns when they have a concrete release artifact and want agent-driven campaign execution instead of ad hoc copy drafts.493installs127Web Performance Optimizationweb-performance-optimization is a davila7/claude-code-templates skill for systematic website and web app performance work. It guides developers to measure, analyze, and improve loading speed, runtime behavior, JavaScript bundle size, and caching strategies with explicit focus on Core Web Vitals metrics LCP, FID, and CLS. The skill also ties performance gains to SEO ranking and conversion outcomes. Developers reach for web-performance-optimization when Lighthouse scores slip, bundles grow unwieldy, pages load slowly on mobile, or agents need a structured checklist to propose concrete frontend perf patches instead of generic advice.485installs128React Best PracticesReact Best Practices is an agent skill that packages a Vercel Engineering performance guide for React and Next.js into actionable rules a developer can invoke while optimizing or reviewing an app. It targets developers who already ship with React 19-era patterns and need a structured checklist instead of scattered blog posts. The skill is meant when you are debugging slow loads, shrinking bundles, refactoring components, or implementing new UI with performance guardrails from day one. Rules span eliminating request waterfalls, cutting bundle size, tuning React Server Components and data fetching, improving client-side caching, and reducing unnecessary re-renders. For Prism’s journey model it shelves under Ship → Performance as the canonical home, with strong overlap in Build → Frontend whenever you write or refactor components. It pairs well with code review and testing skills but does not replace profiling tools or automated Lighthouse runs—you still validate in the browser and CI.477installs129Typescript Experttypescript-expert is a Claude Code skill from davila7/claude-code-templates with 461 installs and skills.sh rank 32 that steers agents toward production TypeScript patterns, strict typing, and scalable architecture. The skill activates during refactors, API typing, generics design, and module-boundary decisions so generated code matches team conventions instead of quick any-typed snippets. Developers reach for typescript-expert when shipping typed Node services, React codebases, or shared libraries where compile-time safety and maintainable interfaces matter more than one-off scripts.477installs130Api Documentation Generatorapi-documentation-generator is a generator skill for developers who ship APIs faster than they write manuals. It walks the agent through analyzing your codebase—routes, HTTP verbs, request and response shapes, auth, and errors—then emits structured endpoint documentation with copy-paste examples and usage guidelines. It is aimed at REST, GraphQL, and WebSocket surfaces when docs are missing, outdated, or blocking onboarding. You invoke it when a feature set is stable enough to describe but you do not want to hand-maintain every parameter table. Output is meant to be developer-friendly and suitable for tightening into OpenAPI or Swagger specifications. It complements code review and security passes but does not replace contract tests or live gateway configuration.474installs131Backend Dev Guidelinesbackend-dev-guidelines is a davila7/claude-code-templates agent skill that encodes reusable backend architecture and coding conventions for Node.js, Python, and Go service development. It helps engineers apply consistent folder structure, error handling, API design, and production-ready patterns instead of improvising layout on every new microservice or monolith module. The skill reports 463 installs and rank 31 on skills.sh from the claude-code-templates source. Reach for it when scaffolding a new backend, reviewing service structure, or aligning multiple language stacks to shared engineering standards inside Claude Code or Cursor agent sessions.473installs132Research Engineerresearch-engineer is an agent skill from davila7/claude-code-templates that adopts a senior research engineer persona focused on scientific rigor, theoretical correctness, and formal verification. The skill bridges theoretical computer science and high-performance implementation by treating requests as peer-reviewed submissions requiring objective criticism rather than agreeable assistance. Developers reach for research-engineer when exploring unfamiliar technical territory, designing benchmarks, planning experiments, or running prototype spikes before committing engineering resources. The workflow emphasizes evidence-backed conclusions and optimal implementation choices across any required technology stack.458installs133Agent Developmentagent-development is a davila7/claude-code-templates skill (version 0.1.0) for building Claude Code plugin agents as markdown files with YAML frontmatter. Required fields are name (3–50 lowercase hyphenated characters) and description with 2–4 `<example>` blocks showing when the agent should trigger; optional model (inherit, sonnet, opus, haiku), color (six UI identifiers), and tools arrays enforce least privilege. The markdown body becomes a second-person system prompt covering responsibilities, analysis steps, output format, and edge cases, kept under 10,000 characters. Agents differ from slash commands: agents run autonomously on matching tasks while commands handle user-initiated actions. Developers reach for agent-development when scaffolding code reviewers, test generators, or domain specialists inside a Claude Code plugin instead of improvising ad-hoc subagent prompts.452installs134Code Reviewcode-review is a davila7/claude-code-templates skill encoding Sentry engineering practices for agent-driven pull request review. Agents scan changes for runtime errors such as null pointer risks and out-of-bounds access, performance problems including unbounded O(n²) loops and N+1 queries, unintended side effects on other components, and backwards compatibility breaks. The skill triggers when reviewing pull requests, examining diffs, or giving code quality feedback across security, performance, testing, and design dimensions. It complements generic review prompts with Sentry-specific expectations rather than subjective nitpicking. Developers reach for code-review when Claude agents must audit Sentry-style Python or backend changes before merge, especially when teams want consistent checklist coverage instead of ad hoc comments on each pull request during ship-phase quality gates and pre-release reviews.447installs135Commitcommit is a Claude Code skill from davila7/claude-code-templates that formats git commits using conventional commits with Sentry-specific issue references. The skill enforces a required header pattern of type, optional scope, and subject under 100 characters, plus body and footer sections. Supported commit types include feat, fix, ref, perf, and docs, each mapped to a clear purpose so agents and developers produce readable history during active feature work. Developers reach for commit when staging changes and needing standardized messages that link work to tracked issues without manually drafting headers and footers.437installs136Data Privacy ComplianceData Privacy Compliance is a skill from davila7/claude-code-templates for data privacy and regulatory compliance across GDPR, CCPA, HIPAA, and international data protection laws. It supports privacy control implementation, Data Protection Impact Assessments, consent management, data minimization, and privacy-by-design reviews before real user data flows through a product. Developers reach for it when drafting privacy policies, mapping retention, enabling data-subject rights, or validating launch readiness for regulated markets. The skill focuses on compliance architecture and process guidance rather than generic security scanning alone.435installs137Git Commit Helpergit commit helper is a davila7/claude-code-templates skill that generates descriptive conventional commit messages by analyzing staged changes. Developers start with git diff --staged, then the agent proposes a message following conventional commits format with clear type and scope. The skill hooks PostToolUse on Bash commands and logs analysis timestamps to ~/.claude/git-commit-helper.log for traceability. Reach for git commit helper when finishing a change set, reviewing staged files, or asking an agent to phrase commits that match the actual diff instead of generic placeholders.435installs138Nowait Reasoning OptimizerGuides optimization of Claude Code agent reasoning so tasks complete faster with fewer redundant planning passes, smarter tool invocation order, and tighter prompt structure for autonomous coding workflows.434installs139Modalmodal is a Claude Code skill from davila7/claude-code-templates that teaches Modal's serverless Python platform for ML inference, GPU training, parallel batch jobs, cron schedules, and HTTPS web endpoints. The SKILL.md walks through modal token new authentication, Image.debian_slim dependency images, @app.function and @app.cls patterns, GPU types from T4 through B200 including multi-GPU H100:8 syntax, Volume persistence, Secret management, and modal deploy for production endpoints. Developers reach for modal when elastic GPU compute, .map() parallelization, or zero-to-thousands autoscaling beats self-managed Kubernetes or fixed VMs. New Modal accounts include $30/month in credits, and the skill bundles 11 reference docs covering images, scaling, volumes, and scheduled jobs.415installs140Accessibility Auditoraccessibility auditor is a web accessibility specialist skill from davila7/claude-code-templates for WCAG compliance, ARIA implementation, and inclusive design review. It audits websites and components for contrast failures, keyboard traps, missing labels, semantic HTML problems, and screen-reader incompatibilities against WCAG 2.1 Level AA or AAA expectations. Developers reach for accessibility auditor before launch when ADA-aligned, assistive-technology-friendly experiences are required and issues must be identified with actionable remediation guidance across keyboard navigation and ARIA patterns.412installs141Manimmanim from davila7/claude-code-templates is a Claude Code skill that guides agents to author Manim Python scenes for explanatory math and concept animations. Manim renders vector motion graphics suited to step-by-step derivations, algorithm visualizations, and narrated technical explainers commonly used in course modules, launch videos, and internal training. The skill helps developers translate mathematical or systems concepts into scripted scenes, camera moves, and timed transitions instead of hand-editing timeline tools. Reach for manim when a project needs reproducible animation source code checked into a repo, batch-rendered clips for documentation sites, or consistent visual language across a tutorial series. It complements static docs and slide decks by turning equations, data flows, and geometric intuition into programmatic video assets developers can re-render after code changes.407installs142Json Canvasjson-canvas from davila7/claude-code-templates is a Claude Code skill for creating and editing valid JSON Canvas files using the .canvas extension per JSON Canvas Spec 1.0. Canvas files are plain JSON with two top-level arrays holding nodes, edges, groups, and connection metadata for infinite-canvas editors such as Obsidian. The skill guides Claude through authoring new canvases, validating structure against the spec, and transforming existing canvas data for mind maps, flowcharts, brainstorm boards, and linked technical documentation. Developers reach for json-canvas when a task mentions .canvas files, Obsidian canvases, visual knowledge maps, or diagram-style docs that must remain machine-readable and portable across compatible tools. The workflow emphasizes spec-compliant JSON rather than proprietary binary formats, so canvases stay diff-friendly in Git and editable outside a single vendor UI.400installs143Email Composeremail composer is a davila7/claude-code-templates agent skill for drafting professional emails across business, technical, and customer communication contexts. It asks for purpose, recipient relationship, key points, and desired tone, then outputs a standard structure with a clear subject line, greeting, context opening, body, call to action, and sign-off. The skill supports requests, follow-ups, announcements, and urgent or friendly variants while keeping copy segmentation-friendly for lifecycle and promotional use. Developers reach for email composer when they need help writing stakeholder updates, customer replies, vendor requests, or onboarding sequences—not when implementing SMTP APIs, Resend integrations, or HTML template engines in code. Quick start requires only the communication context; no external tools or dependencies are declared in the manifest.389installs144Railway Deploymentrailway-deployment is a Claude Code agent skill from davila7/claude-code-templates that operationalizes Railway's CLI for day-two deployment management. Version 1.0.0 depends on railway-cli and restricts agent tools to Bash(railway:*), covering railway deployment list, logs, redeploy, restart, and down workflows rather than initial build setup or service creation. Developers reach for railway-deployment when they need JSON deployment history via deployment list with limits up to 1000, filtered build or runtime logs using @level:error filters and --since time ranges, or --latest logs for failed or in-progress deployments. The skill documents non-interactive redeploy and restart with -y flags and clarifies that railway down stops a running deployment but keeps the service record. For pushing new code, creating services, or deleting infrastructure, the skill routes to composable siblings: railway-deploy, railway-new, railway-environment, and railway-status.389installs145Stripe Integrationstripe-integration is a payments engineering skill from davila7/claude-code-templates, sourced from vibeship-spawner-skills under Apache 2.0, that guides complete Stripe payment system implementation beyond bare API calls. It covers Checkout sessions, subscription lifecycle management, customer billing portal, webhook verification and idempotency, metered billing, Stripe Connect, and dunning for failed payments. The skill emphasizes edge cases that break revenue at scale: declined cards, webhook delivery failures, subscription state transitions, and 3am billing incidents. Developers reach for stripe-integration when adding payments, subscriptions, billing, or checkout to a SaaS or ecommerce backend and need production-grade error handling rather than tutorial-level charge creation. Output includes webhook endpoints, checkout flows, subscription schemas, and billing portal integration code. It assumes Stripe as the payment processor and does not cover alternative processors or tax compliance outside Stripe's built-in tools.389installs146Error Resolvererror-resolver in davila7/claude-code-templates helps developers triage broken builds and runtime failures without sprawling refactors. The skill analyzes stack traces, failing test output, and runtime exceptions to isolate root causes and suggest minimal code changes plus verification steps that confirm the fix. Agents use it when CI turns red, local test suites fail, or production logs surface uncaught exceptions that block shipping. Developers reach for error-resolver during incident triage or pre-release hardening when time pressure favors targeted patches over architectural rewrites. The workflow emphasizes proposing the smallest change that restores green builds and healthy production behavior, then listing concrete checks to run before merge. Part of the claude-code-templates collection, error-resolver pairs naturally with test and lint skills but stays focused on active failure recovery rather than preventive review. Triggers include fix this stack trace, failing test diagnosis, runtime exception debugging, and restore CI green status.386installs147Task Execution Enginetask-execution-engine is a Claude Code skill from davila7/claude-code-templates focused on building dependable execution layers for autonomous agents. It guides developers through decomposing complex coding or operations work into tracked steps, persisting state between steps, retrying transient failures, and surfacing blockers instead of silently halving a plan. The skill targets engineers who outgrew one-shot prompts and need a runner that can resume after errors, parallelize safe subtasks, and report progress across long refactors or deployment sequences. Teams reach for task-execution-engine when agents drop mid-migration, lose context between shell commands, or cannot explain which step failed in a ten-part workflow. It complements individual domain skills by supplying orchestration patterns—queues, checkpoints, and retry policy—rather than domain-specific commands alone.386installs148Codex Reviewcodex-review is an agent skill integrated with the Codex CLI for professional code review on diffs, branches, and pull requests before commits or merges. The skill flags bugs, regressions, and style issues during large-scale refactors and automatically generates CHANGELOG entries from reviewed changes. Install via `npx skills add -g BenedictKing/codex-review`, ensure Codex CLI is present, then invoke with `/codex-review` or natural-language triggers. Developers reach for codex-review when they want AI-assisted review and changelog drafting without manually writing release notes.374installs149It Operationsit-operations is an agent skill from davila7/claude-code-templates for day-two IT operations after services are live. The skill covers incident triage, uptime and endpoint health checks, on-call runbook execution, scheduled change windows, and steady-state hygiene across production services. Developers and platform engineers reach for it-operations when alerts fire, SLAs slip, or routine ops tasks—failover checks, runbook steps, post-incident cleanup—need structured agent guidance instead of ad-hoc shell commands.373installs150Codexcodex is an AI & Agent Building skill from davila7/claude-code-templates that templates Codex-style agent behavior for Claude Code sessions. The skill sets up repo-aware editing patterns, terminal command execution, automated test runs, and iterative fix loops so agents can propose changes, validate them, and self-correct within a single workflow. Developers reach for codex when they want OpenAI Codex-like autonomy—reading the codebase, running tests in shell, and looping on failures—without manually re-prompting each iteration. The template fits teams standardizing agent configuration across repositories that already use Claude Code as their primary coding agent.366installs151Accessibilityaccessibility is an agent skill from davila7/claude-code-templates (metadata version 1.0) that guides WCAG 2.1 Level A, AA, and AAA conformance checks for public web and mobile UIs. It inspects semantic HTML, keyboard tab order, focus management, color contrast ratios, ARIA landmarks, alt text, and screen-reader labels aligned with Lighthouse and axe DevTools patterns. Manual testing guidance covers NVDA, JAWS, and VoiceOver checks plus 200% zoom reflow validation. The skill returns prioritized critical, serious, and moderate remediation steps with concrete HTML and ARIA code fixes. Reach for accessibility before launch when legal WCAG AA compliance, ADA readiness, or user-reported keyboard and screen-reader failures block release.344installs152Move Code Qualitymove-code-quality is a davila7/claude-code-templates skill that reviews Move packages against the official Move Book Code Quality Checklist using Move 2024 Edition standards. It activates on `.move` files or `Move.toml` manifests and scans 11 categories covering 50+ specific rules—from package edition declarations to module syntax, struct design, option macros, testing, and comments. Critical checks include requiring `edition = "2024.beta"` or `edition = "2024"` in Move.toml and flagging legacy patterns incompatible with modern Sui development. Output separates critical issues, important improvements, and recommended enhancements with file paths, line numbers, and fix snippets. Developers reach for move-code-quality before merging Move modules or preparing Sui smart contract audits.340installs153Zapier Workflowszapier-workflows is a Claude Code template skill from davila7/claude-code-templates for planning Zapier automation. It helps developers map triggers, filters, and actions across SaaS tools, produce clear Zap documentation, and avoid redundant custom webhook code when Zapier can orchestrate the flow. Reach for it when wiring CRM updates, notification pipelines, form-to-database flows, or multi-app handoffs during MVP or internal tooling work. The skill emphasizes readable workflow specs agents and teammates can implement directly in the Zapier editor.335installs154Railway Deployrailway-deploy is a Claude Code template skill from davila7/claude-code-templates that guides developers through deploying applications to Railway with production-ready service configuration. The skill covers defining Railway services, setting environment variables, specifying build and start commands, configuring health checks, and sequencing release steps so hosted apps start reliably after push or CI triggers. Developers reach for railway-deploy when moving a Node, Python, or containerized API or web app from local dev to Railway without guessing platform-specific settings. The workflow emphasizes repeatable deploys: explicit build commands, secret management through Railway env vars, health endpoint verification, and rollback-friendly release ordering. Use it during pre-launch hardening or when migrating an existing stack to Railway hosting.334installs155Figmafigma is a design-handoff skill for frontend developers implementing UI from Figma sources. It instructs agents to pull layout grids, color and spacing tokens, typography scales, and component hierarchy from Figma files so React, Vue, or HTML implementations match the design system. Developers invoke figma when a ticket includes a Figma URL and they need token lists, spacing values, and component boundaries before writing CSS or component code. The skill reduces guesswork in padding, font stacks, and nested frame structure—bridging design files and codebase conventions from the davila7/claude-code-templates collection.329installs156Pyhealthpyhealth is a davila7/claude-code-templates scientific skill for the PyHealth healthcare AI library, not generic API health probes. The skill documents a five-stage pipeline—data loading, task definition, model selection, training, and deployment validation—across 10+ datasets such as MIMIC-III/IV, eICU, and OMOP, 20+ predefined clinical tasks, and 33+ models including RETAIN, SafeDrug, and Transformer. Six reference files (~28,000 words) cover datasets, medical coding, tasks, models, preprocessing, and training evaluation, with split_by_patient guidance to prevent leakage. Developers reach for pyhealth when building mortality prediction, hospital readmission, drug recommendation, sleep staging, or ICD coding workflows with PyTorch Trainer checkpoints and clinical fairness or calibration checks.329installs157Create Prcreate-pr is a davila7/claude-code-templates skill that opens pull requests following Sentry engineering conventions. The workflow verifies branch state with git status and git log main..HEAD, analyzes the full commit and diff scope, and drafts titles plus descriptions with testing notes for reviewers. The skill requires an authenticated GitHub CLI (gh) binary. Developers reach for create-pr when a feature branch is ready and they want consistent PR structure instead of manual gh pr create typing. Outputs are merge-ready PR bodies aligned to Sentry code review guidelines.316installs158Railway Databaserailway-database is a Claude Code skill for developers standing up backend data stores on Railway without manual dashboard guesswork. It walks through provisioning Railway-managed Postgres or compatible databases, setting environment variables, and producing connection strings apps can consume from local and deployed environments. The workflow targets SaaS and API backends that need a managed database online quickly during feature development. Use railway-database when starting a new service, replacing a local-only database, or connecting an existing app repo to Railway-hosted persistence.313installs159Tdd Workflowtdd-workflow is a skill from davila7/claude-code-templates that encodes Test-Driven Development principles for Claude Code agents. It follows the RED → GREEN → REFACTOR loop and the three TDD laws: production code only to pass failing tests, minimal test code to show failure, and minimal implementation to go green. Allowed tools include Read, Write, Edit, Glob, Grep, and Bash so agents can create tests, implement code, and run suites quickly. Developers reach for tdd-workflow when they want agent-enforced TDD instead of bolting tests on after features land.295installs160Torch Geometrictorch-geometric is an agent skill in davila7/claude-code-templates for PyTorch Geometric graph neural network development on irregular relational data. The SKILL.md covers Data objects with x, edge_index, edge_attr, and y tensors, benchmark Planetoid loaders such as Cora, and 40+ convolutional layers including GCN, GAT, and GraphSAGE variants documented in references/layers_reference.md. Training sections address node classification, graph classification, link prediction, heterogeneous HeteroData graphs, and large-scale NeighborLoader sampling. Bundled references include datasets_reference.md, transforms_reference.md, api_patterns.md, and a layer_capabilities cheatsheet. Installation examples use uv pip install torch_geometric with optional pyg_lib wheels from data.pyg.org. Developers reach for torch-geometric when agents must convert CSV or numpy edge tables into PyG pipelines, pick conv layers, or debug edge_index COO orientation during GNN training.292installs161Railway Statusrailway-status is a davila7/claude-code-templates agent skill that checks live Railway deployment health for the current directory using railway status --json. It verifies the Railway CLI is installed, confirms railway login authentication, detects linked projects, and parses JSON for project name, workspace, environment, per-service deployment state, activeDeployments build progress, and configured domains. The skill is scoped as a pre-flight gate before deploys or variable changes and explicitly defers environment edits to the railway-environment skill. Reach for railway-status when an incident, release, or support ticket needs confirmation whether failures are platform, build, or runtime related. Skip it when you need to change Railway env vars, redeploy code, or configure services—use railway-deploy or railway-environment instead.291installs162Agent Managementagent-management is part of the AI Maestro skill suite in davila7/claude-code-templates, providing CLI-driven lifecycle control for multiple AI coding agents. Through aimaestro-agent.sh commands, developers create agents with project directories and task tags, list and inspect agent status, hibernate agents to save resources, wake them with optional tmux attach, rename, export and import configurations, and install or uninstall plugins per agent. Agent statuses track online, offline, and hibernated states across tmux sessions. The skill requires AI Maestro running locally with tmux 3.0 plus and sits alongside five sibling skills for messaging, memory search, docs search, graph queries, and planning within the six-skill AI Maestro orchestration platform. Reach for agent-management when running parallel backend and frontend agents, pausing overnight work without losing state, or installing specialized plugins on individual agent instances.290installs163Cocoindexcocoindex is a Claude Code skill from davila7/claude-code-templates that guides developers through the CocoIndex real-time data transformation framework for AI workloads. The skill walks through defining Python indexing flows that ingest from five supported source types—local files, S3, Azure Blob, Google Drive, and Postgres—apply built-in transforms such as text chunking, SentenceTransformer embeddings, and LLM extraction, then export to six targets including Postgres+pgvector, Qdrant, LanceDB, Neo4j, and Kuzu. CocoIndex's incremental processing model reprocesses only changed records and supports live sync from sources to indexes. Developers reach for cocoindex when standing up document embedding pipelines, knowledge graph builders, or low-latency retrieval layers for RAG agents. The skill covers dependency setup, custom `@coco.fn` functions, CLI and Python API operation, and flow patterns with collectors and top-level exports documented at cocoindex.io.289installs164Railway Newrailway-new is a deployment bootstrap skill in davila7/claude-code-templates that guides Claude Code through creating a fresh Railway project for apps, APIs, and background workers. It automates linking a Git repository, defining environments, and pointing services at Railway deploy targets so developers skip repetitive dashboard clicks. Teams reach for railway-new when prototyping a Node, Python, or containerized service that should go live on Railway directly from an agent session. The skill fits the claude-code-templates collection aimed at repeatable infrastructure recipes rather than one-off shell snippets.286installs165Railway Servicerailway-service is a deployment skill for Claude Code that guides developers through standing up and configuring applications on Railway PaaS. The skill covers service creation, environment variable setup, custom domain binding, and health check configuration for backends, background workers, and full-stack apps heading to production. It is built for engineers who already have runnable code and need Railway-specific wiring instead of generic Docker or Kubernetes steps. Use railway-service when you are ready to ship a service to Railway and need env vars, domains, and health endpoints configured correctly the first time.286installs166Using Git Worktreesusing-git-worktrees is a Claude Code skill from davila7/claude-code-templates for spinning up isolated git worktrees before feature work or plan execution. The workflow follows a priority order for directory selection, verifies repository safety, and announces worktree setup at session start so agents do not overwrite in-progress changes on the main branch. Developers reach for using-git-worktrees when they need two or more Claude Code sessions on different branches simultaneously, when implementation plans require a clean sandbox, or when frequent git checkout would risk losing uncommitted work. The skill encodes systematic isolation as directory selection plus verification rather than ad-hoc folder copies.286installs167Railway Projectsrailway-projects is a Claude Code skill for listing, organizing, and managing multiple Railway projects and environments from the agent. The skill helps developers who operate several microservices, staging stacks, or client-specific Railway deployments keep projects discoverable and environments separated. Developers reach for railway-projects when Railway sprawl makes it hard to track which service lives in which project or which environment maps to which client. The skill complements per-project configuration skills by focusing on portfolio-level Railway organization rather than single-environment variable wiring.284installs168Railway Templatesrailway-templates is a version 1.0.0 MIT agent skill from davila7/claude-code-templates that searches and deploys services from Railway's template marketplace using railway-cli and a GraphQL helper script. Allowed tools are Bash scoped to railway:* commands. The skill documents common template codes such as ghost, strapi, n8n, minio, and uptime-kuma, while directing database-only adds to the companion railway-database skill. Workflows fetch verified templates, read serializedConfig JSON, and deploy into a target Railway project and environment. Developers invoke railway-templates when they need CMS, automation, storage, or monitoring services without browsing the Railway UI. Prerequisites include railway status for project and environment IDs plus workspace context from the Railway API.284installs169Railway Environmentrailway-environment is a Claude Code skill for configuring Railway projects so applications run consistently across staging and production. The skill guides developers through environment variables, attached services, database provisioning, custom domains, and deploy hook setup on Railway’s platform. Developers reach for railway-environment when a Railway app needs environment parity, secret management, or production-ready service wiring without manually hunting Railway dashboard settings. The skill fits teams shipping Node, Python, or containerized APIs and frontends on Railway who want agent-assisted infrastructure configuration rather than ad hoc CLI trial and error.283installs170Railway Domainrailway-domain is a version 1.0.0 MIT agent skill from davila7/claude-code-templates authored for Railway. It wraps railway-cli commands to generate Railway-provided domains, add custom domains, inspect current URLs, and remove domains from deployed services. Generate a default subdomain with railway domain --json, optionally targeting a service via --service, noting max one Railway domain per service and requiring an existing deployment. Custom domains use railway domain example.com --json and return required DNS records for verification. Domain state lives under config.services.*.networking with serviceDomains and customDomains arrays. Removal and reads compose with the railway-environment skill to stage and apply networking changes. Developers reach for railway-domain when they need a public URL, branded hostname, or DNS setup after Railway deploy.278installs171Railway Metricsrailway-metrics is a version 1.0.0 agent skill from davila7/claude-code-templates that queries Railway resource telemetry through a GraphQL metrics API. After obtaining environmentId and optional serviceId from railway status --json, agents request measurements such as CPU_USAGE, CPU_LIMIT, MEMORY_USAGE_GB, MEMORY_LIMIT_GB, NETWORK_RX_GB, NETWORK_TX_GB, DISK_USAGE_GB, EPHEMERAL_DISK_USAGE_GB, and BACKUP_USAGE_GB over configurable ISO 8601 time windows with sampleRateSeconds and averagingWindowSeconds. groupBy supports deploymentInstanceId, deploymentId, serviceId, and region tags. Developers reach for railway-metrics when answering how much memory a service uses, whether CPU is saturated, or if network or disk growth signals a regression. Combine with railway-service for broader health checks beyond raw resource charts.278installs172Tavily Webtavily-web is a davila7/claude-code-templates skill for Tavily-powered web search, content extraction, crawling, and research. Install with npx skills add -g BenedictKing/tavily-web, configure a Tavily API key, then use natural language in Claude Code for grounded retrieval and citation-friendly results. Developers reach for it when agents must answer with current web data, pull page content from URLs, or crawl sites instead of relying on stale model knowledge. Examples and patterns live in the BenedictKing/tavily-web GitHub repository referenced by the skill guide.264installs173Iterate Priterate-pr is a pull request iteration skill from davila7/claude-code-templates that automates the feedback-fix-push-wait cycle until CI is green. It requires an authenticated GitHub CLI and starts by identifying the current branch PR with gh pr view, then checks GitHub Actions status before reading review threads. Developers reach for iterate-pr when failing checks or reviewer comments block merge and they want continuous fixes instead of hand-polling Actions. The skill tightens diffs, updates PR descriptions, and pushes follow-up commits until review feedback and CI gates clear.249installs174Pdf Processingpdf-processing is a document-handling skill from davila7/claude-code-templates that guides Python and command-line workflows for reading, creating, merging, splitting, rotating, watermarking, encrypting, and OCR-processing PDF files. It covers pypdf for basic operations, pdfplumber for text and table extraction to Excel, reportlab for PDF generation, plus qpdf, pdftotext, and pytesseract for CLI and scanned-document tasks. Developers reach for pdf-processing when a feature touches .pdf uploads, invoice parsing, form filling, or compliance document pipelines. Companion references FORMS.md and REFERENCE.md cover advanced form fill and JavaScript pdf-lib patterns. The quick-reference table maps eight common tasks to the best library or command.246installs175Postgres Schema Designpostgres-schema-design is a comprehensive PostgreSQL table design reference from davila7/claude-code-templates covering data types, indexing, constraints, performance patterns, and advanced Postgres features. Core rules recommend PRIMARY KEY on reference tables with BIGINT GENERATED ALWAYS AS IDENTITY by default, UUID only when global uniqueness or opacity is required, and normalization to 3NF before measured denormalization for proven read hotspots. Developers reach for postgres-schema-design when starting new SaaS entities, API persistence layers, or ecommerce catalogs that need durable, migration-friendly Postgres schemas. The skill emphasizes eliminating redundancy and update anomalies while documenting when denormalization earns its maintenance cost.245installs176Medchemmedchem is a cheminformatics prototyping skill from davila7/claude-code-templates for developers exploring medicinal-chemistry workflows in agent-assisted sessions. It helps draft scaffold filtering steps, molecular property prediction experiments, and structure-activity relationship (SAR) exploration before investing in a production cheminformatics stack or lab-facing tooling. Use medchem when the goal is to validate feasibility—compound filters, ADMET proxies, or SAR tables—in code sketches rather than run regulated lab software. The skill suits Claude Code or Cursor users in biotech, pharma tooling, or research engineering who need fast iteration on chemistry logic. Reach for medchem when exploring pipelines, not when shipping validated LIMS integrations.244installs177Adaptyvadaptyv is a claude-code-templates skill for embedding mobile subscription infrastructure into iOS and Android applications. It guides developers through Adapty-style SDK integration patterns covering paywall presentation, entitlement checks, free-trial configuration, restore-purchases handling, and subscription analytics hooks across platform-specific build targets. The skill helps agents place SDK initialization, customer identification, product catalog wiring, and receipt validation in the correct app lifecycle points without relearning store billing APIs from scratch. Developers reach for adaptyv when shipping freemium or premium mobile features that require cross-platform subscription state, trial conversion tracking, and compliant restore flows on App Store and Google Play builds. It assumes an existing mobile codebase rather than greenfield backend services.243installs178Template Skilltemplate-skill is a Claude Code agent skill scaffold from davila7/claude-code-templates that bootstraps a new skill with the correct file structure, trigger description, and instruction sections contributors expect in a shared templates repository. The bundled SKILL.md frontmatter includes name and description placeholders—"Replace with description of the skill and when Claude should use it"—plus a heading where authors insert operational instructions below. Developers reach for template-skill when publishing a new agent capability and want consistent formatting across many skills instead of copying ad-hoc markdown from older repos. The skill accelerates contributor onboarding in claude-code-templates by standardizing how triggers are written and how body sections are organized. Use it at skill creation time before filling in domain-specific workflows, checklists, or integration steps.239installs179Heygen Best PracticesGuides Claude through HeyGen best practices for AI-generated presenter videos: planning scripts, choosing avatars and voices, batching variants, and exporting assets suited for landing pages, social posts, and nurture campaigns.227installs180PyopenmspyOpenMS skill supports building proteomics and metabolomics backends that parse mass-spec files, run peak detection, and quantify features. It targets researchers and engineers wiring scientific Python pipelines into APIs or CLI batch jobs.227installs181Security Best PracticesThe security-best-practices skill steers Claude through established application security patterns for SaaS, APIs, and agent workflows. It helps teams catch unsafe defaults, tighten authentication and secrets handling, and satisfy review gates before shipping production-facing software.219installs182Playwright E2e BuilderPlaywright E2E builder skill helps teams create reliable browser automation for SaaS and web clients. It structures specs, fixtures, and assertions while aligning with CI pipelines so regressions are caught before users see broken flows.218installs183Exa SearchExa-search integrates Exa's semantic search to find high-quality web pages, product references, and technical sources for early ideation, competitive scans, and agent grounding with attributable citations instead of stale training recall.211installs184Github Actions CreatorCreates production-ready GitHub Actions pipelines with test jobs, caching, environment secrets, and merge gates so repositories automate quality checks and controlled releases from pull request to deploy.199installs185PlanningStructured planning skill that turns ambiguous goals into scoped milestones, task lists, dependencies, and clear acceptance criteria so implementation starts with aligned priorities and realistic delivery boundaries.185installs186PlaywrightInstructs Claude on Playwright end-to-end testing: resilient selectors, network stubbing, auth flows, parallel CI execution, and diagnostic artifacts so web apps gain dependable regression coverage before production deploys.181installs187Footballbin PredictionsFootballbin-predictions skill helps Claude build quick football forecasting prototypes: ingest match statistics, apply prediction heuristics or models, format confidence scores, and demo analytics content for validating a sports data product concept.175installs188Fastmcp ServerFastmcp-server accelerates building Model Context Protocol servers with FastMCP conventions for defining tools, resources, prompts, authentication, and deployment so LLM agents can invoke your backend capabilities through a standardized integration layer.147installs189Memory SearchGuides Claude to design memory-search for AI agents: ingestion, embeddings, filters, query APIs, and grounding rules so assistants retrieve the right prior context without hallucinating or leaking unrelated session data.141installs190Design To CodeDesign-to-code skill guides Claude to convert visual designs into clean frontend implementations: map frames to components, honor design tokens, preserve UX states, and deliver accessible responsive UI for web or hybrid product surfaces.134installs191Using NeonGuides Claude Code through Neon serverless PostgreSQL setup including project creation, branch databases per environment, secure connection strings, and integration patterns for modern SaaS and API backends requiring scalable relational storage.108installs192Agent MessagingAgent-messaging skill equips Claude to architect clear communication between agents: define payloads, sequence handoffs, handle async replies, and keep multi-agent pipelines traceable for production agent systems and internal automations.104installs193Screenshotscreenshot is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted development.102installs194Docs SearchQueries the AI Maestro documentation index semantically or by keyword to look up functions, classes, and API docs before implementing changes. A developer uses it to verify correct signatures and patterns against indexed codebase docs.99installs195Rust Cli BuilderInterview-driven workflow for building Rust command-line tools with clap, subcommands, config support, and distribution planning. A developer uses it when scaffolding a new Rust CLI before writing code.97installs196Vercel Deployvercel-deploy is a Claude Code skill for devops & ci/cd. It helps solo builders move faster with AI-assisted development.97installs197Neon InstagresInstantly provisions a Neon Postgres database with pooler and direct connection URLs for local development. A developer uses it to spin up a dev or production database in seconds and wire it into an ORM.95installs198Commit SmartInspects the working tree and diff to suggest staging and generate conventional commit messages with the right type and scope. A developer uses it to produce meaningful, context-rich commits from their actual changes.92installs199Verl Rl TrainingProvides production RL training guidance with verl, supporting RLHF/GRPO/PPO at scale with swappable FSDP, Megatron, vLLM, and SGLang backends. A developer uses it for scalable LLM post-training including agentic and vision-language RL.85installs200Distributed Llm Pretraining TorchtitanProvides torchtitan-based large-scale LLM pretraining with composable FSDP2, tensor, pipeline, and context parallelism plus Float8 and torch.compile. A developer uses it to pretrain Llama 3.1, DeepSeek V3, or custom models at multi-GPU scale.83installs201Jupyter Notebookjupyter-notebook is a Claude Code skill for productivity & planning. It helps solo builders move faster with AI-assisted development.83installs202Security Threat Modelsecurity-threat-model is a Claude Code skill for security. It helps solo builders move faster with AI-assisted development.82installs203Torchforge Rl TrainingProvides RL research guidance with torchforge, a PyTorch-native library that isolates algorithms from infrastructure via Monarch and TorchTitan. A developer uses it for clean RL abstractions and scalable experimentation.82installs204Fastapi EndpointAdds production-ready FastAPI endpoints with async SQLAlchemy queries, Pydantic v2 validation, and generated pytest coverage. A developer uses it to build CRUD or authenticated API routes inside an existing FastAPI project.78installs205Miles Rl TrainingProvides guidance for large-scale RL post-training with miles, a production fork of slime, targeting MoE models with FP8/INT4 and speculative RL. A developer uses it when training 1TB+ MoE models needing train-inference alignment.72installs206Slime Rl TrainingProvides RL post-training guidance with slime, tightly integrating Megatron-LM and SGLang for GLM models and custom data workflows. A developer uses it when scaling RL training with Megatron integration.69installs207Graph QueryQueries the AI Maestro code graph database to reveal component relationships, call chains, and change impact. A developer uses it after reading code to understand what depends on a component before modifying it.68installs208Transcribetranscribe is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted development.63installs209Test DetectDetects the testing framework from project files and runs the appropriate test command for all tests, a specific file, or generates basic tests. A developer uses it to run the right tests without knowing the setup.42installs210X Twitter ScraperGuides building X (Twitter) integrations through the Xquik API covering tweet search, user lookups, follower extraction, monitoring, and webhooks. A developer uses it when adding X data access to an app. Requires an Xquik API key.35installs211GccStructures agent memory into a Git-like versioned file system to persist milestones, explore alternatives in branches, and recover context across sessions. A developer uses it on long multi-step projects needing durable, structured memory.34installs212Claude Apiclaude-api is a Claude Code skill for backend & apis. It helps solo builders move faster with AI-assisted development.31installs213Tailwind Design SystemCovers building Tailwind CSS design systems with tokens, component variants, theming, and accessible responsive patterns. A developer uses it when creating a component library or standardizing UI in Tailwind.31installs214Api Security TestingA phased security-testing workflow for APIs that enumerates endpoints and tests auth, JWT/OAuth flows, rate limiting, and input validation. A developer or bug-bounty tester uses it to assess REST and GraphQL endpoint security.29installs215Gmod Addon MakerGuides Garry's Mod addon development including Lua scripting across Server/Client/Shared realms, asset creation, and packaging for distribution. A developer uses it when building, debugging, or organizing GMod addons.28installs216Data ScientistProvides the full data-science workflow from exploratory analysis to production models, spanning statistics, ML algorithms, and business insight. A developer uses it for statistical analysis, experiment design, forecasting, and predictive modeling.26installs217Python ProProvides expert Python guidance covering modern 3.12+ features, async workflows, and production tooling. A developer uses it when building or reviewing production Python services.22installs218Agirails Agent PaymentsAn interactive onboarding workflow that generates customized agent payment code for ACTP escrow, x402 payments, and USDC settlement on Base L2. A developer uses it to integrate agent-to-agent payments with proper key management and the ACTP state machine.21installs219Think TankSimulates a moderated multi-persona expert debate to surface trade-offs and blind spots before planning begins. A developer uses it when making architecture, technology, or strategy decisions where multiple perspectives help.21installs220Architecture PatternsGuides selecting and applying proven backend architecture patterns, defining module boundaries, dependency rules, and migration steps. A developer uses it when designing new systems, decoupling monoliths, or planning microservices decomposition.20installs221Premium Web DesignProduces premium, agency-quality website and landing-page designs as React components that avoid generic AI aesthetics. A developer uses it when a site should look expensive, editorial, or award-winning.19installs222Zod Validation ExpertCovers Zod schema validation including parsing, custom errors, refinements, type inference, and ecosystem integrations. A developer uses it when validating API inputs, forms, or environment variables in TypeScript.19installs223Database ArchitectGuides greenfield and re-architecture data-layer design including technology selection, schema modeling, indexing, partitioning, and migration planning. A developer uses it when designing a scalable database architecture from day one.18installs224Web Security TestingA phased workflow for testing web applications against the OWASP Top 10, from reconnaissance through vulnerability validation. A developer uses it for pen testing, security assessments, or bug bounty work.18installs225Cf CrawlInitiates Cloudflare Browser Rendering crawl jobs, polls to completion, and writes page content as markdown for local project context. A developer uses it to ingest documentation sites or knowledge bases into their repo.17installs226Data EngineerGuides designing and implementing data pipelines, warehouses, and lakehouse architectures with ingestion, transformation, validation, and governance. A developer uses it to build reliable, cost-effective analytics infrastructure.17installs227PostgresqlGuides PostgreSQL-specific schema design covering data types, constraints, indexing, partitions, and RLS policies. A developer uses it when modeling a Postgres schema for scale and maintainability.17installs228Tanstack Query ExpertProvides production patterns for TanStack Query as an async state manager including caching, mutations, and Next.js SSR hydration. A developer uses it when replacing useEffect fetching or building a data layer.17installs229Typescript ProCovers advanced TypeScript typing including generics, conditional/mapped types, strict compiler flags, and decorators for production systems. A developer uses it when hardening type safety or solving complex typing problems.17installs230Code Review ExcellenceGuides reviewing pull requests for correctness, security, performance, and maintainability, producing feedback grouped by severity with rationale. A developer uses it to run better reviews, set review standards, or mentor through feedback.16installs231Fastapi ProProvides expert guidance for building async-first FastAPI APIs, microservices, and real-time WebSocket features with modern Python patterns. A developer uses it when designing scalable, production-ready FastAPI services.16installs232Postgresql OptimizationProvides a PostgreSQL optimization workflow spanning query tuning, indexing, vacuum management, and config tuning. A developer uses it when addressing slow queries or administering production databases.16installs233Sql ProCovers modern SQL across cloud-native and hybrid OLTP/OLAP systems with performance tuning and data modeling. A developer uses it when optimizing queries or designing database schemas.16installs234Api Design PrinciplesGuides API design decisions across REST and GraphQL, including resource modeling, error handling, versioning, pagination, and auth strategy. A developer uses it when designing new APIs, refactoring existing ones, or setting team API standards.15installs235Database OptimizerProvides query optimization, EXPLAIN ANALYZE plan analysis, indexing, and caching guidance across PostgreSQL, MySQL, SQL Server, Oracle, and NoSQL. A developer uses it to eliminate bottlenecks and speed up slow queries.15installs236Domain Driven DesignGuides DDD work with a viability check, strategic artifacts (subdomains, bounded contexts, glossary), and routing to CQRS, event sourcing, and saga patterns. A developer uses it when modeling a complex business domain with explicit boundaries.15installs237Github Actions TemplatesSupplies reusable GitHub Actions workflow patterns for CI/CD across various stacks, including test, build, deploy, and security-scan jobs. A developer uses it when automating testing and deployment pipelines.15installs238Javascript ProProvides expert guidance on modern JavaScript and async programming across Node.js and browsers. A developer uses it when debugging async behavior or migrating legacy JS to modern ES standards.15installs239Nodejs Backend PatternsSupplies production-ready Node.js backend architecture patterns spanning APIs, microservices, auth, and real-time features. A developer uses it when building scalable, maintainable Node.js backends.15installs240React Component PerformanceIdentifies React render hotspots, isolates expensive updates, and applies memoization and prop stabilization. A developer uses it to profile and speed up slow React components.15installs241Security AuditA workflow bundle for comprehensive security audits across web apps, APIs, penetration testing, and hardening. A developer uses it to run a structured security assessment of an application.15installs242Code Simplifiercode-simplifier is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted development.14installs243Github AutomationManages GitHub issues, PRs, branches, and deployments programmatically through Composio's GitHub toolkit over Rube MCP. A developer uses it to script repository and code-review workflows.14installs244Ml EngineerProvides expert ML engineering guidance for production systems including model serving, feature engineering, and ML infrastructure. A developer uses it when building scalable, monitored ML systems.14installs245Php ProProvides expert guidance on writing idiomatic, high-performance PHP with generators, iterators, SPL structures, and modern OOP. A developer uses it when building or reviewing performance-sensitive PHP code.14installs246React State ManagementCovers React state patterns spanning local, global, server, URL, and form state with library selection criteria and quick-start code. A developer uses it when setting up global state, managing server state, or migrating from legacy Redux.14installs247Gitops Workflowgitops-workflow is a Claude Code skill for automation & workflows. It helps solo builders move faster with AI-assisted development.13installs248Nextjs App Router PatternsSupplies Next.js 14+ App Router architecture patterns covering Server Components, streaming, and Server Actions. A developer uses it when building App Router apps or migrating from Pages Router.13installs249ScrapeFetches webpage content as clean markdown through Bright Data's Web Unlocker API with automatic bot and CAPTCHA bypass. A developer uses it when scraping pages that block ordinary requests. Requires BRIGHTDATA_API_KEY and unlocker zone.13installs250Brightdata Local SearchSets up the local unfancy-search pipeline on top of Bright Data's SERP API, adding query expansion, multi-engine retrieval, RRF reranking, and dedup. A developer uses it to run a self-hosted web search engine locally via Docker or Node.12installs251Javascript Testing PatternsProvides testing patterns for JS/TS applications spanning unit, integration, and end-to-end tests with mocking and TDD. A developer uses it when setting up test infrastructure or writing tests for components and APIs.12installs252Performance OptimizerDiagnoses and fixes performance bottlenecks across code, databases, and APIs with a measure-optimize-verify process. A developer uses it when an app, page, query, or endpoint is slow.12installs253Python Testing PatternsProvides Python testing patterns using pytest, fixtures, and mocking across unit, integration, and async tests with TDD. A developer uses it when building test suites or writing Python tests.12installs254React Native ArchitectureSupplies production-ready React Native and Expo architecture patterns covering navigation, state, native modules, and offline-first design. A developer uses it when starting or scaling a React Native mobile app.12installs255Saas Multi TenantCovers multi-tenant SaaS data architecture using PostgreSQL row-level security, tenant isolation, and cross-tenant admin patterns. A developer uses it when building a shared-schema multi-tenant backend.12installs256Bright Data McpDirects the agent to route all web search, scraping, and structured data extraction through the Bright Data MCP server, which handles JavaScript, CAPTCHAs, and bot detection. A developer uses it for web research, e-commerce and social data, and browser automation instead of built-in web tools.11installs257Claude Code SessionsReads Claude Code's session JSONL files to make past sessions searchable and analyzable, offering 11 skills plus a localhost dashboard. A developer uses it to find past work, check token usage, resume sessions, or manage orphaned tasks.11installs258Database MigrationCovers schema and data migrations across major Node ORMs with rollback procedures and zero-downtime deployment techniques. A developer uses it when migrating between ORMs or databases, refactoring data models, or upgrading schema versions.11installs259DiagrammerConverts a plain-English diagram request into a JSON node-and-edge spec and renders it to an SVG via the local diagrammer tool. A developer uses it to produce precise technical diagrams for docs, READMEs, posts, or slides without a design tool.11installs260Flutter ExpertProvides expert Flutter guidance for multi-platform apps, covering widget composition, Impeller rendering, and platform-specific integrations from a unified codebase. A developer uses it when building or optimizing Flutter apps across mobile, web, and desktop.11installs261Kubernetes ArchitectProvides expert Kubernetes architecture guidance covering cluster topology, GitOps, service mesh, and multi-tenancy at scale. A developer uses it when designing K8s platforms or implementing GitOps delivery.11installs262N8n Workflow PatternsSupplies reusable n8n workflow architecture patterns derived from real usage, covering webhooks, API integration, and database sync. A developer uses it when designing reliable n8n automation flows.11installs263Prompt Engineering PatternsSupplies advanced prompt-engineering patterns to improve LLM output performance, consistency, and control. A developer uses it when designing or optimizing prompts for production LLM applications.11installs264Zustand Store TsScaffolds Zustand stores from a template with typed state, separated actions, and subscribeWithSelector middleware. A developer uses it when adding a Zustand store to a TypeScript React app.11installs265Bash ProGuides writing hardened Bash for automation and CI/CD using strict mode, safe argument parsing, and defensive patterns, tested with Bats and ShellCheck. A developer uses it when writing or reviewing shell scripts for ops safety and portability.10installs266Cloud DevopsA phased workflow bundle for cloud infrastructure setup, container orchestration, CI/CD, monitoring, and cost management across major clouds. A developer uses it to provision infrastructure and stand up cloud-native deployments end to end.10installs267Design MirrorScreenshots and scrapes an inspiration site through Bright Data, extracts its design system, and translates those tokens into the user's own framework and components. A developer uses it to make their app match the aesthetic of a referenced URL.10installs268Docx OfficialSkill for generating DOCX documents from templates. Professional document creation.10installs269Golang ProProvides expert Go guidance covering modern language features, concurrency patterns, and scalable microservice architecture. A developer uses it when building or reviewing Go services, CLIs, or high-performance systems.10installs270Laravel ExpertGuides production-grade Laravel development emphasizing clean architecture, testability, security, and performance on Laravel 10/11+. A developer uses it when building or reviewing idiomatic Laravel applications.10installs271Deep Research NotebooklmRuns a scoped research workflow through the NotebookLM MCP for market, competitive, prospect, and trend analysis, producing briefs plus optional artifacts. A developer uses it to research a topic deeply and generate slides, audio, videos, or reports.9installs272Observability EngineerDesigns production observability covering monitoring, logging, tracing, and SLI/SLO-driven alerting. A developer uses it when instrumenting services or investigating reliability regressions.9installs273Secrets Managementsecrets-management is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted development.9installs274AngularProvides expert guidance for building and modernizing Angular v20+ apps with Signals, standalone components, zoneless configuration, and SSR. A developer uses it when creating new Angular apps or migrating away from NgModules to modern reactive patterns.8installs275Async Python PatternsProvides asyncio concurrency patterns including tasks, gather, queues, pools, cancellation, timeouts, and backpressure for I/O-bound Python. A developer uses it when building async APIs, scrapers, real-time services, or concurrent I/O workloads.8installs276Data FeedsFetches parsed structured data from major websites via Bright Data's Web Data APIs by passing a URL and dataset type, with automatic polling. A developer uses it to pull product details, profiles, reviews, and posts as clean JSON without scraping logic.8installs277Deadline PrepCombines the .claude change-log CSV with today's git history to categorize work and produce talking points for a demo. A developer uses it to quickly prep a standup or deadline demo from what they actually shipped.8installs278Rust ProProvides Rust expertise in async programming, advanced types, and production systems programming. A developer uses it when building performance-critical or systems-level Rust applications.8installs279SearchRuns Google searches through Bright Data's SERP API and returns structured JSON results. A developer uses it to add programmatic web search to an app or agent. Requires BRIGHTDATA_API_KEY and unlocker zone.8installs280Supply Chain GuardScans npm, PyPI, crates.io, GitHub Actions, and CI/CD pipelines for known compromised packages, malicious versions, filesystem IOCs, and C2 indicators. A developer uses it to detect and remediate supply chain attacks.8installs281Bright Data Best Practicesbright-data-best-practices is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted development.7installs282Debugging StrategiesTurns debugging into methodical problem-solving via reproduction, hypothesis formation, binary search, and targeted instrumentation. A developer uses it to track down hard bugs, performance issues, or production incidents.7installs283Incident ResponderProvides an SRE incident-response playbook covering rapid triage, incident command, and post-incident review. A developer uses it during production outages to resolve and communicate incidents systematically.7installs284K6 Load TestingBuilds realistic k6 load-test scenarios (smoke, load, stress, spike, soak) and analyzes results to validate performance. A developer uses it to catch performance regressions and check SLA compliance before deployment.7installs285Llm OpsCovers production LLM operations such as RAG pipelines, embeddings with Pinecone/Chroma/pgvector, fine-tuning, evals, and cost control. A developer uses it when building production AI architectures.7installs286Monorepo ArchitectProvides expert guidance on monorepo tooling, workspace configuration, build caching, and dependency-graph management. A developer uses it when setting up or scaling a monorepo across teams.7installs287Playwright JavaProduces enterprise-grade Playwright Java E2E tests with Page Object Model, JUnit 5, Allure reporting, and thread-safe parallel runs. A developer uses it when writing or enhancing Java browser test automation.7installs288Terraform SpecialistCovers advanced Terraform/OpenTofu IaC including module design, remote state backends, GitOps, and policy-as-code for multi-cloud stacks. A developer uses it when designing or automating infrastructure as code.7installs289C ProProvides C systems-programming guidance covering memory management, pointers, POSIX system calls, pthreads, and debugging with valgrind and gdb. A developer uses it when writing performance-sensitive or embedded C code.6installs290DebuggerA debugging specialist that captures the error and stack trace, reproduces and isolates the failure, and implements a minimal verified fix. A developer uses it proactively when hitting errors or test failures.6installs291Dependabot ReviewLists open Dependabot PRs, classifies each into risk tiers from branch name and title, and auto-merges safe patches while reporting major or security bumps. A developer uses it to triage and clear dependency-update PRs.6installs292Dotnet BackendProvides expert guidance for building ASP.NET Core 8+ APIs with EF Core, auth (Identity, JWT, OAuth), background workers, and validation. A developer uses it to build or refactor .NET backend services and data access.6installs293E2e Testing PatternsGuides building E2E test suites around critical user journeys with stable selectors, test data strategies, tracing, and isolation. A developer uses it to implement or stabilize end-to-end tests and wire them into CI.6installs294Java ProProvides expert Java guidance covering modern JVM features and Spring Boot 3.x for production enterprise applications. A developer uses it when building or reviewing Java 21+ services.6installs295Jira AutomationManages Jira issues, projects, sprints, and boards programmatically through Composio's Jira toolkit over Rube MCP. A developer uses it to script issue tracking and project workflows.6installs296Linear AutomationManages Linear issues, projects, and cycles programmatically through Composio's Linear toolkit over Rube MCP. A developer uses it to script issue tracking and project workflows.6installs297Llm EvaluationProvides comprehensive LLM evaluation strategies spanning automated metrics, human review, and A/B testing. A developer uses it to measure LLM app performance, compare models or prompts, and catch regressions before deployment.6installs298Pydantic AiGuides building production AI agents with PydanticAI, giving typed, validated LLM outputs, tool use, and testable dependency injection across providers. A developer uses it when building Python agents needing correctness and testability.6installs299Best PracticesDocumentation templates for coding best practices using Claude Code. Educational reference.1installs300InstructorEducational skill providing instructor resources and teaching templates. Learning support.1installs301Isms Audit ExpertISMS Audit Expert is listed in davila7’s Claude Code templates as a security-focused agent skill aimed at Information Security Management System audits—the kind of structured control review aligned with ISO 27001 thinking. Prism’s ingest shows placeholder asset and reference readme sections rather than a complete procedural SKILL.md, so treat this entry as a catalog stub you may need to flesh out locally before production audits. For founders selling B2B SaaS, the intended use is to walk gaps in policies, risk treatment, access control evidence, and logging retention with consistent questioning, then map answers to control objectives. It is phase-specific to Ship/security because audits gate launch and enterprise deals more than day-one coding. Confidence is moderated until real checklist steps and output formats appear in the repository.1installs302Session Handoffsession-handoff is a Claude Code skill for ai & agent building. It helps developers move faster with AI-assisted coding.1installs303Webapp Testingwebapp-testing is a Claude Code skill for testing & qa. It helps developers move faster with AI-assisted coding.1installs304AccessibilityAudit and improve web accessibility following WCAG 2.1 guidelines. Use when asked to "improve accessibility", "a11y audit", "WCAG compliance", "screen reader support", "keyboard navigation", or "make ...0installs305Chembl DatabaseQuery ChEMBL's bioactive molecules and drug discovery data. Search compounds by structure/properties, retrieve bioactivity data (IC50, Ki), find inhibitors, perform SAR studies, for medicinal chemistr0installs306Star History Chartstar-history-chart is a Claude Code skill from davila7/claude-code-templates that puts a 'Stargazers over time' chart in a GitHub README without relying on any third-party chart service. GitHub now restricts the /stargazers endpoint to a repo's own admins and collaborators, which broke star-history.com's free tier, starchart.cc and every similar live-image service with a 'Requires authentication' error. The skill's fix is structural: a bundled Python script fetches stargazer timestamps with an authenticated token and renders a clean, light/dark-adaptive SVG committed to the repo, and a bundled GitHub Actions workflow regenerates docs/star-history.svg on a weekly cron (plus manual trigger) using the repo's built-in GITHUB_TOKEN - so there are no secrets to configure and nothing external to break. Point the README at the local SVG once and the chart stays live forever.0installs