nimadorostkar/Claude-Skills-collection

Use when designing or reviewing an HTTP API. Covers resource modeling, status codes, pagination, idempotency, versioning, error formats, and the contract details that break clients when you get them wrong.

Provides patterns for designing reliable LLM agents with tool use, including loop termination, error handling, human checkpoints, and recognizing when an agent isn't needed.

Write short, accurate instruction files for coding agents, including commands, conventions, and hard constraints. Use it when setting up or refreshing CLAUDE.md, AGENTS.md, or similar files.

A method for agents to persist critical state—facts, failures, conclusions—in a file that survives context compaction and session ends, with strategies for retrieval and preventing stale data.

Architecture Decisions records technical choices as ADRs with context, options, the decision, and consequences. Use it when a significant or hard-to-reverse decision needs to be documented for future maintainers.

Use when designing asynchronous job processing. Covers queue selection, idempotency, retry and backoff policy, scheduling, poison messages, and observability for work that happens outside the request.

Brand Guidelines helps document and apply a brand system across new work. Use it to extract colours, type, spacing, imagery, and voice from existing assets, then audit for drift and consistency.

Bug Fix Protocol guides a disciplined bug fix workflow end to end: reproduce the issue, write a failing test, apply the smallest fix, verify it, and check for related cases. Use it for reported defects, regressions, and hotfixes.

Use when adding or debugging a cache. It covers cache placement, invalidation, stampede protection, TTL selection, and the consistency tradeoffs involved.

Competitive Analysis gathers evidence on competitors, pricing, positioning, and differentiation to inform a product decision. It is used when entering a market, repositioning, or responding to a competitor move.

Manage what an LLM sees by budgeting the context window, retrieving relevant information, compacting history, and pruning tool results to prevent attention degradation.

Data Modeling helps design database schemas for new tables or major changes. It covers normalization, keys, temporal history, multi-tenancy, soft deletes, and constraints that are hard to reverse later.

Data Quality catches bad data before it reaches a dashboard, model, or customer. It profiles datasets, validates schema and business rules, checks freshness and volume, and fails the pipeline with quarantine and alerts when checks break.

Database Performance helps diagnose slow queries, lock contention, and connection saturation using execution plans and query logs. Use it when the database is the bottleneck or you need to make schema and index changes for speed.

Build Django applications with efficient ORM usage, secure settings, and safe migrations. Use when developing or reviewing Django apps, APIs, or database designs.

Use when designing systems around events and message queues. Covers event schema design, delivery guarantees, idempotent consumers, ordering, dead-letter handling, and event sourcing.

FastAPI builds APIs with dependency injection, Pydantic validation, async database access, authentication, background tasks, and tests. Use it when building or reviewing FastAPI services and structuring handlers and dependencies.

Guides the decision to fine-tune versus prompt engineering or retrieval, and provides a rigorous process for dataset construction, training with LoRA or full fine-tuning, and evaluation to achieve reliable performance improvements.

Generative Art creates algorithmic visuals with seeded randomness, noise, palettes, and composition rules. Use it when building reproducible procedural art or data-driven visual work.

GraphQL helps design and operate GraphQL APIs with schema modeling, resolver batching, query cost limits, error handling, and federation. Use it when building public or internal graphs and diagnosing slow or expensive queries.

Enforce rules deterministically in AI agents using lifecycle hooks. Automate formatting, block dangerous commands, and inject context without depending on model memory.

Optimize LLM feature costs by identifying where tokens are spent and applying caching, context reduction, model routing, batching, and output limits without degrading output quality.

Provides a systematic approach for measuring LLM feature quality, including building evaluation sets, selecting metrics, using LLM-as-judge, regression testing, and production evaluation. Use before iterating on prompts or models to avoid guesswork.

Use when integrating an LLM API into an application. Covers streaming, retries and rate limits, timeouts, caching, fallback across providers, and the production concerns that a tutorial integration ignores.

Lead Research qualifies accounts, finds trigger events, and identifies the right contact before outreach. Use it when you need a brief with one specific observation that makes outreach relevant instead of generic.

Use when building a Model Context Protocol server. Covers tool, resource, and prompt design, transport choice, authentication, error handling, and testing against a real client.

Build and operate reproducible machine learning pipelines. Covers feature engineering, train/serve consistency, drift monitoring, and retraining to avoid silent model degradation.

Microservices helps decide whether to split a system into services and how to set the boundaries. It covers service communication, distributed data, sagas, and the operational costs of networked calls.

Helps select the right LLM for a task by evaluating capability, cost, latency, and routing strategies. Use when choosing a model for a new feature, reducing costs, or evaluating model migrations.

NestJS helps you build NestJS services with clear module boundaries, validation pipes, guards, interceptors, and data access patterns. Use it when structuring a service or writing unit and end-to-end tests.

Pandas helps analyze and transform tabular data in Python with vectorized operations, explicit dtypes, and correct joins. Use it when cleaning data, reviewing pandas code, or deciding if the data is too large for pandas.

Packages skills, commands, hooks, and MCP servers into an installable plugin. Use it when creating manifests, bundling capabilities, testing clean installs, or publishing updates.

PostgreSQL helps you tune schemas, indexes, autovacuum, pooling, and replication for Postgres workloads. Use it when diagnosing slow queries, bloat, locks, or planning large migrations.

Product Analysis helps assess product performance and decide what to build using metrics, funnels, retention, and feature adoption data. It is used to separate signal from noise and turn analysis into a prioritized recommendation.

Write and improve prompts for language models. Covers instruction design, few-shot examples, reasoning elicitation, output formatting, and systematic failure diagnosis.

Use when building retrieval-augmented generation. Covers chunking, embedding and hybrid search, reranking, grounding and citation, and diagnosing whether a bad answer is a retrieval failure or a generation failure.

Realtime WebSockets helps build live features with WebSockets or SSE. Use it to choose a protocol, handle reconnects and backfill, and scale message delivery across instances.

Resume helps write or tailor a CV that shows outcomes instead of responsibilities. Use it to restructure bullets, quantify impact, and adapt to a role without inventing experience.

A guide for creating skills that trigger reliably and change an AI agent's behavior. Use when writing or debugging a skill, focusing on scoping, description writing, and progressive disclosure.

Skill Name helps with a specific workflow by defining when to use it, what inputs it needs, and the steps to follow. Use it when you need a repeatable process with clear outputs and guardrails.

Use when reviewing or improving an existing agent skill. Covers triggering accuracy, content quality, redundancy with the base model, and measuring whether the skill actually changes behavior.

Provides guidance for creating reusable slash commands for AI agents, covering argument handling, workflow composition, and distinguishing commands from skills.

Spark helps you build and review distributed Spark jobs with a focus on partitioning, shuffles, skew, joins, caching, and Spark UI diagnosis. Use it when a pipeline is slow, spilling, or has a straggling task.

Spring Boot helps build and review Spring Boot services with correct transaction boundaries, JPA fetching, configuration validation, and test slices. Use it when diagnosing lazy loading, N+1 queries, or slow full-context tests.

Use when an LLM must return machine-readable data. It covers schema design, native structured-output modes, validation and repair, and extraction from messy input.

Design and dispatch subagents for context isolation, parallel work, or verification. Use when delegating tasks to a separate agent with its own context is more efficient than processing in the main context.

Theming builds readable light, dark, and high-contrast themes with semantic tokens, contrast checks, and palette generation from a seed colour. Use it when adding dark mode, multi-brand themes, or fixing contrast across theme variants.

UI Review critiques interface designs by checking hierarchy, affordance, consistency, and state coverage. Use it when reviewing mockups, screenshots, or live screens and you need feedback that leads to specific design fixes.

User Research helps plan interviews, usability tests, and synthesis into findings grounded in observed behavior. Use it when you need evidence for a product decision, not opinions or feature wish lists.

Visual Design produces static layouts like posters, covers, and social graphics with deliberate hierarchy, composition, type, and colour. Use it when reviewing or creating designs for print or screen.