Stripe integrations can fail when teams choose outdated APIs or unclear payment flows. This skill guides Claude, Codex, and Claude Code toward current Stripe patterns.

Skill authors need clear structure, strong triggers, and reusable resources. This skill gives Claude and Codex a practical workflow for designing, writing, validating, and improving Claude Code plugin skills.

Plugin authors need a reliable layout before adding commands, agents, skills, hooks, or MCP servers. This skill explains the structure and manifest patterns that make Claude Code plugins portable and maintainable.

Plugin developers need consistent per-project settings without committing user state. This skill shows how to use local markdown files with YAML frontmatter, validation, and defaults.

Plugin developers need reliable MCP setup guidance for external tools and services. This skill explains server types, authentication, tool naming, and testing steps for Claude Code plugins.

Claude Code hooks can be hard to design safely because they run during tool events. This skill gives patterns, examples, and scripts for validation, testing, and policy enforcement.

Large feature work often slows down when developers do not know where code belongs. This skill maps the DevPrep AI folder structure and recommends precise locations for new code.

Teams lose the reasons behind important technical choices, which causes repeated debates and inconsistent implementation. This skill creates structured ADRs with context, decisions, consequences, alternatives, and review guidance.

API teams often lose time resolving inconsistent routes, errors, versions, and documentation. This skill provides practical patterns, checklists, and templates for REST, GraphQL, gRPC, OpenAPI, and AsyncAPI work.

AI applications need reliable retrieval, tool use, cost controls, and monitoring to work in production. This skill provides patterns and templates for RAG, agents, vector databases, streaming, and observability.

Researchers often need to connect SRA, GEO, BioProject, and publication records before analysis. This skill guides Claude, Codex, or Claude Code through SRAgent commands for accession conversion, metadata inspection, and paper retrieval.

Blocklet release steps can fail when versioning, build output, or metadata are misaligned. This skill guides Claude, Codex, and Claude Code through a repeatable release workflow with validation before bundling.

Packaging a web project as a blocklet requires several ArcBlock files and validation steps. This skill guides conversion for static sites and Next.js apps using a provided DID.

Large refactors often change behavior by accident. This skill guides Claude, Codex, and Claude Code through one tested structural change at a time.

Code reviews can miss outdated dependencies, changed documentation, and project-specific rules. This skill guides Claude, Codex, and Claude Code through a deeper review with current ecosystem checks.

Software bugs often get worse when agents guess at fixes too early. This skill enforces a structured investigation, hypothesis, test, and verification workflow.

Generic interfaces often feel forgettable and disconnected from their product context. This skill helps Claude, Codex, and Claude Code create polished frontend work with a clear visual direction.

Writing a skill from scratch can be unclear for new plugin authors. This skill gives a compact reference for structure, frontmatter, descriptions, and skill content.

Financial modeling requires consistent assumptions, valuation logic, and sensitivity testing. This skill helps Claude, Codex, and Claude Code create structured DCF models and scenario analyses.

Cookbook notebooks need consistent review before publication. This skill combines automated notebook checks with a structured rubric for clear, actionable audit reports.