Coding standards, frameworks, refactoring, and code review (18.4K skills)
Design firmware and peripheral drivers for ARM Cortex-M boards such as Teensy, STM32, nRF52, and SAMD. The skill helps with interrupts, DMA, memory and cache behavior, and maintainable C++ or Rust code for embedded systems.
Turn component documentation into a deployment view of a system. Map components to containers such as applications, databases, and queues, then document their interfaces, dependencies, infrastructure needs, and communication in Mermaid diagrams and API specifications.
Turn code-level notes into a component view of a system. Define each component’s responsibility, boundaries, interfaces, and relationships, then produce consistent documentation and Mermaid diagrams that show how components within a container interact.
Design backend services and APIs by defining service boundaries, contracts, and communication patterns. Get guidance on REST, GraphQL, or gRPC, resilience, security, observability, testing, and rollout planning for systems that need to scale and remain maintainable.
Plan or refactor backend systems using Clean Architecture, Hexagonal Architecture, and Domain-Driven Design. Define domain boundaries, modules, interfaces, and dependency rules, then outline migration steps and checks for maintainable, testable code.
Review system designs and major code changes for architectural risks and tradeoffs. Get guidance on service boundaries, scalability, resilience, security, and maintainability, along with concrete improvement ideas and validation steps.
Create and review Bash scripts for automation, CI/CD, and system tasks. Apply safer input handling, error reporting, portability practices, and testing with tools such as Bats and ShellCheck to make scripts easier to run and maintain.
Find application-wide performance bottlenecks using profiling and observability data, then guide backend and frontend improvements, load testing, and ongoing monitoring. It helps teams measure changes against baselines and set safeguards against regressions.
Analyze an existing codebase from its directories upward, then document code, components, deployment containers, and system context using the C4 model. Produce linked Markdown documentation, Mermaid diagrams, and OpenAPI specifications in a dedicated directory.
Design asynchronous Python systems for APIs and I/O-heavy work such as network or database calls. Choose task, queue, or pool patterns, and account for timeouts, cancellation, backpressure, errors, testing, and debugging.
Configure Bazel for large monorepos, investigate dependencies, and improve build performance. The skill provides patterns for build files, caching, remote execution, custom rules, and profiling so teams can diagnose slow builds and shape a more efficient build setup.
Review code for duplication, complexity, and design problems, then plan and apply small refactoring steps while keeping behavior stable. The result is clearer, easier to maintain code with verification notes.
Plan and carry out AngularJS-to-Angular migrations, from choosing a rewrite or gradual hybrid approach to updating components, services, routing, and forms. Use it to assess risks, organize milestones, and validate an incremental transition.
Use this skill to inspect compiled programs by reading assembly, mapping branches and loops, and recognizing common code patterns. It helps you infer functions, variables, data structures, and program logic from executable files using static analysis.
Implement, resume, review, or verify the DeepSeek++ scheduled automation feature. Follow its GitHub issues and progress records while preserving the design that schedules runs in the background and executes DeepSeek requests in the page context.
Plan and carry out large project transformations by analyzing the codebase, breaking work into phases, and setting up documentation and progress tracking. It can use GitHub Issues when available and resume work from existing project records.
Review code changes in a working tree, a commit date range, or a branch diff. It prioritizes evidence-backed findings such as bugs, regressions, correctness risks, missing tests, and security or data-safety defects, with findings ordered by severity.
Provides a minimal starting point for writing a new skill, with fields for its name and description and a place for instructions. The supplied template contains no working task instructions.
Convert product or technical specifications in Notion into linked implementation plans and actionable tasks. The workflow extracts requirements, dependencies, and acceptance criteria, then tracks task status and progress so a development effort can be followed from planning through completion.
Build MCP servers, which give AI agents tools for external services, using Python or TypeScript. The skill guides tool design, input validation, error handling, code review, and evaluations based on realistic tasks.
Turn Git commits, records of code changes, into clear release notes for users. The skill groups updates by type, rewrites technical descriptions in plain language, and filters out internal work such as refactoring and tests.
Create or improve Claude skills that package specialized knowledge, repeatable workflows, and tool instructions. The guide covers planning SKILL.md and supporting files, generating a starter folder, validating the structure, and packaging the result as a distributable ZIP.
Build complex claude.ai artifacts with React, Tailwind CSS, and shadcn/ui. Create interfaces with multiple components, state management for changing data, and routing between views, then package the app into one self-contained HTML file for sharing.
Get parallel feedback on a request from three delegates using models from different providers. The skill gathers their findings, summarizes agreements and disagreements, and suggests next steps; code reviews also include a score from each delegate.