Coding standards, frameworks, refactoring, and code review (15.2K skills)
Build or review Django applications using examples for project structure, APIs, database access, and caching. Apply patterns for separating settings and business logic, validating requests, and reducing repeated database queries.
Design and review error handling in TypeScript, Python, and Go applications. Use structured error types, controlled retries, API error responses, and readable failure messages to make failures visible and separate user guidance from developer diagnostics.
Design task-specific workflows for repeated agent work, with pass/fail checks, progress records, and handoff notes. Identify when a temporary workflow should become a shared skill that teammates can run and evaluate again.
Add background jobs to Django with Celery, moving work such as email delivery and PDF generation outside web requests. Covers scheduled runs, retries, linked or parallel tasks, monitoring, and tests, with Redis or RabbitMQ carrying task messages.
Prevent token balances and values from being scaled incorrectly across EVM blockchains. Look up each token's decimal places at runtime, cache them by chain and address, and use precise arithmetic with consistent units.
Find the ECC command sequence for a development task, including the order to run commands and when to stop. It reads current command files, groups related workflows, and provides guidance without executing them.
Coordinate parallel AI agent sessions with dmux, a tool that manages terminal panes. Split development, research, testing, or review into separate tasks, isolate overlapping file changes, and review results before merging them.
Coordinate parallel frontend, backend, or service development around one shared, machine-checkable interface specification. Generate consumer types, validate mocks and real responses against the same specification, and review changes together to prevent mismatched fields and incompatible integrations.
Review a feature or proposal from four perspectives: product management, architecture, development, and quality assurance. Get each role’s concerns and first actions, followed by a summary of shared priorities and unresolved disagreements.
Look up current library and framework documentation through Context7 MCP, a connection to live documentation tools. Supports setup questions, API references, and code examples, using version-specific sources when available to ground answers in retrieved material.
Write, review, and refactor C# and .NET code using practical conventions for asynchronous work, configuration, and dependency injection—passing services into classes. Includes examples for data access and APIs to support maintainable applications.
Write, review, and refactor modern C++ using the C++ Core Guidelines. Covers type safety, automatic resource cleanup, class design, exceptions, concurrency, and naming conventions, with examples and a checklist for identifying unsafe or inconsistent code.
Create CodeTour files that guide readers through a codebase step by step, with verified file and line references. Tailor the walkthrough to new contributors, architecture readers, PR reviewers, or investigators tracing a failure or security boundary.
Diagnose failed AI agent runs by capturing errors, checking likely causes, and choosing small, reversible recovery actions. The workflow produces a structured report of the failure, recovery steps, and evidence, helping explain repeated retries or stalled progress.
Investigate unreliable AI agents across 12 layers, including instructions, memory, tool execution, and display. Use code and logs to trace failures, rank findings by severity, and produce an ordered repair plan that prioritizes code changes.
Build or review server-side APIs in Node.js, Express, and Next.js. Apply patterns for organizing business logic and data access, reducing unnecessary database queries, caching results, validating requests, and handling authentication and errors.
Build a project-specific ECC installation plan using evidence from code and configuration. Sort skills, rules, hooks, and other components into items loaded every session and references kept available for occasional use.
Integrate x402, an HTTP payment protocol, so agents can pay for APIs or services. Configure task and session budgets, approved recipients, and wallets with privately managed keys, using integration paths for Base, X Layer, or Solana.
Explore an unfamiliar codebase and turn its structure, request flow, and development conventions into a concise onboarding guide. Create or update CLAUDE.md with project-specific instructions, preserving existing guidance and identifying findings that remain uncertain.
Use Bun to run JavaScript and TypeScript, manage project dependencies, build applications, and run tests. Includes guidance for choosing between Bun and Node, migrating existing projects, and configuring deployments on Vercel.
Turn speedup requests into a measured optimization process with limits on time, cost, and variants. Compare each change against a baseline and correctness checks, then retain the fastest safe option with reproducible commands and measurements.
Design caching for repeated PDF parsing, text extraction, or image analysis using SHA-256 hashes, identifiers calculated from file contents. Reuse results after files move or change names, and process them again when their contents change.
Record software design decisions as numbered Markdown files, with context, rejected alternatives, benefits, drawbacks, and risks. Maintain an index and consult existing ADRs—architecture decision records—to explain why a framework, database, or design approach was chosen.
Coordinate coding projects through Claude DevFleet by splitting work into dependent tasks and dispatching Claude Code agents in separate Git working directories. Monitor progress, manage automatic task sequencing, and review structured reports of changes, tests, and errors.