Avinandan gupta
Community@whatswithavi
just a normal coder
Agent Skills by Avinandan gupta
Showing 59 vetted skills indexed across 1 GitHub repositories.
whatswithaviclone-website
Reverse-engineers websites into pixel-perfect Next.js clones using browser automation and parallel builder agents.
whatswithaviplanning-with-files
Organizes complex multi-step tasks using persistent markdown planning files on disk.
whatswithaviui-ux-pro-max
Searches a BM25-indexed database of UI styles, palettes, typography, and stack guidelines for design decisions.
whatswithaviusing-git-worktrees
Creates isolated git worktrees with directory selection, ignore verification, and baseline test validation.
whatswithavimulti-agent-patterns
Designs and implements multi-agent architectures with supervisor, swarm, and hierarchical coordination patterns.
whatswithavidefuddle
Extract clean markdown content from web pages using the Defuddle CLI.
whatswithavibdi-mental-states
Transform RDF context into BDI agent mental states using formal ontology patterns.
whatswithavicontext-degradation
Diagnose and mitigate context degradation patterns in LLM agent systems.
whatswithavitest-driven-development
Enforces red-green-refactor test-driven development workflow before writing implementation code.
whatswithavisystematic-debugging
Diagnose bugs through a four-phase root cause investigation process before proposing fixes.
whatswithavitheme-factory
Apply curated color palettes and font pairings to slides, docs, and HTML artifacts.
whatswithavicontext-compression
Implement structured summarization and probe-based evaluation for compressing long agent conversation histories.
whatswithavimemory-systems
Designs agent memory architectures using vector stores, knowledge graphs, and temporal retrieval frameworks.
whatswithavidoc-coauthoring
Guides collaborative document drafting through context gathering, iterative refinement, and reader testing.
whatswithaviweb-design-guidelines
Review UI code files against Vercel Web Interface Guidelines and report findings.
whatswithaviadvanced-evaluation
Implements LLM-as-judge evaluation pipelines with direct scoring, pairwise comparison, and bias mitigation.
whatswithaviobsidian-markdown
Create and edit Obsidian Flavored Markdown with wikilinks, embeds, callouts, and properties.
whatswithaviusing-superpowers
Enforces skill discovery and invocation before any AI response or action.
whatswithavidispatching-parallel-agents
Dispatches concurrent agents to investigate independent test failures across separate problem domains.
whatswithaviclaude-api
Build LLM-powered applications with the Claude API and Anthropic SDKs.
whatswithavixlsx
Create, edit, and recalculate Excel spreadsheets with formulas using openpyxl and LibreOffice.
whatswithaviExtract, merge, split, create, and fill PDF documents using Python libraries and command-line tools.
whatswithaviexecuting-plans
Executes written implementation plans task-by-task with verification checkpoints and review gates.
whatswithavialgorithmic-art
Generates p5.js generative art with seeded randomness and interactive parameter controls.
Frequently Asked Questions About Avinandan gupta
FAQPage SchemaWhat tasks can I accomplish with whatswithavi's skill registry?▼
You can clone websites, plan multi-step projects with file-based tracking, run test-driven development and systematic debugging, process PDF/Word/Excel/PowerPoint files, design React and mobile UIs, optimize Postgres queries, build MCP servers, and engineer agent memory, context, and evaluation systems.
Who is the target audience for these skills?▼
Software engineers, agent-system architects, and frontend developers working inside Claude Code-style environments. The manifest suits practitioners building multi-agent systems, enforcing TDD and code-review discipline, designing React/Next.js interfaces, or managing Obsidian vaults and document pipelines.
How do the planning and execution skills work in practice?▼
planning-with-files creates task_plan.md, findings.md, and progress.md, with hooks that reload state after session clears. writing-plans drafts implementation steps, executing-plans runs them with review checkpoints, and verification-before-completion demands command output evidence before claiming success.
Are these skills open source and what licenses apply?▼
Licensing is mixed. Vercel-authored skills (react-best-practices, composition-patterns, react-native-skills) and supabase-postgres-best-practices are MIT. Document skills (pdf, docx, xlsx, pptx) are proprietary, while others reference complete terms in bundled LICENSE.txt files.
What prerequisites or dependencies do these skills require?▼
Most skills run inside a Claude Code environment with Read, Write, Edit, Bash, Glob, and Grep access. Specific ones need extras: webapp-testing requires Playwright, defuddle uses the Defuddle CLI, obsidian-cli needs an Obsidian vault, and planning-with-files invokes PowerShell or shell check scripts.