Claude Skills by ssrjkk

Build reliable LLM agents and agentic workflows: tool use, memory, loops, guardrails, and evaluation. Use for autonomous AI systems.

Designs and curates few-shot examples to guide LLM behavior, including example selection, formatting, and ordering. Use for task specification without fine-tuning.

Implement robust LLM function/tool calling: schemas, multi-call handling, validation, error recovery, and structured execution. Use for agent tool use.

Build evaluation suites for LLM applications: golden datasets, metrics, LLM-as-judge, regression gates, and CI integration. Use for trustworthy model behavior.

Fine-tunes open-source LLMs (Llama, Mistral, Qwen) using LoRA/QLoRA with HuggingFace and Unsloth. Use for domain-specific model adaptation.

Build Model Context Protocol servers and clients: tools, resources, prompts, transport, and secure agent integration. Use for connecting agents to external systems.

Operate machine learning in production: experiment tracking, pipelines, model registry, serving, monitoring, and CI/CD for ML. Use for ML lifecycle.

Optimize LLM cost and latency with prompt caching: cacheable prefixes, cache-control headers, context layout, and cache-aware prompting. Use for high-volume apps.

Design effective prompts for LLMs: role framing, structured output, chain-of-thought, few-shot, and evaluation. Use to get reliable model behavior.

Build retrieval-augmented generation pipelines: chunking, embeddings, vector search, hybrid retrieval, and citation. Use for grounded LLM answers over your data.

Deno runtime and standard library

Build secure web applications with Django: models, views, ORM, admin, auth, REST APIs, and deployment. Use for Python web backends.

Build Node.js web APIs with Express: routing, middleware, error handling, validation, and production hardening. Use for Node backends.

Build high-performance Python APIs with FastAPI: routing, Pydantic validation, async, dependency injection, OpenAPI, and testing. Use for any Python backend.

Build production REST APIs in Go with the standard library or Gin, including routing, middleware, JSON handling, and testing. Use for performant backends.

Build high-performance services with gRPC: protobuf schemas, unary/streaming RPCs, interceptors, and error handling. Use for inter-service communication.

Creates Node.js server-side applications with NestJS, modules, dependency injection, and decorators. Use for enterprise-grade Node.js APIs.

Design consistent REST APIs: resource modeling, status codes, pagination, versioning, error contracts, and documentation. Use for any API design task.

Async Rust with Tokio runtime

Zero-knowledge proof development

Design and operate Elasticsearch: mappings, indexing, queries, aggregations, and cluster tuning. Use for search and log analytics.

Analyze and transform tabular data with pandas: dataframes, cleaning, aggregation, joins, and time series. Use for any data analysis or ETL task.

Write correct and efficient SQL: joins, aggregations, window functions, CTEs, and query optimization. Use for any relational data access.

Design and operate MongoDB: document modeling, queries, indexes, aggregation, replication, and sharding. Use for flexible NoSQL data.

Design and operate PostgreSQL databases: schema design, indexing, query optimization, transactions, and migrations. Use for any relational data layer.

Models databases and writes type-safe queries with Prisma ORM. Use for modern Node.js/TypeScript database access.

Use Redis for caching, sessions, queues, rate limiting, and pub/sub: data structures, persistence, eviction, and clustering. Use for any high-performance data layer.

Design and operate vector databases for semantic search and RAG: embeddings, indexes (HNSW), filtering, hybrid search, and scaling. Use for AI retrieval.

Builds cross-platform desktop applications with Electron, React, and IPC communication. Use for native desktop apps with web tech.

Builds and deploys serverless functions with AWS Lambda, API Gateway, and SAM/CDK. Use for event-driven architectures.

Cloud-native AI deployment patterns

Containerize applications with Docker: Dockerfiles, images, networking, volumes, compose, and production hardening. Use for any deployable service.

Automate CI/CD with GitHub Actions: workflows, jobs, matrices, caching, artifacts, and reusable workflows. Use for build, test, and deploy pipelines.

Configures GitLab CI/CD pipelines with stages, jobs, and GitLab Runner. Use for Git-native automation and deployment.

Design and operate Kafka event streaming: topics, producers, consumers, consumer groups, partitioning, and exactly-once. Use for event-driven systems.

Deploy and operate applications on Kubernetes: workloads, services, config, scaling, and GitOps. Use for container orchestration at scale.

LLM observability with Langfuse/LangSmith

Platform engineering with Backstage/Port

Set up monitoring with Prometheus and Grafana: metrics, exporters, alerting rules, dashboards, and SLO tracking. Use for observability.

Serverless AI inference (Cloudflare Workers, Lambda)

SRE SLI/SLO/SLA implementation

Provision infrastructure as code with Terraform: resources, modules, state, workspaces, and remote backends. Use for cloud resource management.

Rust for embedded systems

AI-powered test generation and validation

Use Claude Code effectively: slash commands, context management, hooks, MCP, permissions, and workflows. Use for agentic coding with Claude Code.

Conduct effective code reviews: correctness, security, performance, style, and actionable feedback. Use for reviewing any pull request or diff.

Use Git effectively: branching, commits, rebase vs merge, history hygiene, conflict resolution, and collaboration workflows. Use for any git-based project.

Bun runtime for JavaScript/TypeScript

Build full-stack React applications with Next.js App Router, Server Components, API routes, middleware, and deployment. Use for production web apps.

Build modern user interfaces with React 19, including Server Components, Actions, hooks, and the new compiler. Use for any interactive UI work.