Technology-agnostic blueprint generator for creating comprehensive copilot-instructions.md files that guide GitHub Copilot to produce code consistent with project standards, architecture patterns, and exact technology versions by analyzing existing codebase patterns and avoiding assumptions.
This skill is a technology-agnostic blueprint generator for producing a comprehensive copilot-instructions.md file that guides GitHub Copilot to generate code consistent with a project's standards, architecture, and exact technology versions. It addresses the problem of Copilot introducing patterns, APIs, or language features that do not match a project by grounding all guidance in the codebase's actual, observed patterns rather than assumptions.
The skill exposes configuration variables for project type, architecture style (layered, microservices, monolithic, domain-driven, event-driven, serverless, or mixed), code-quality focus (maintainability, performance, security, accessibility, testability, or all), documentation level, testing requirements, and versioning scheme. From these it assembles a structured instructions document that establishes priority guidelines, technology-version detection rules (never using features beyond detected language, framework, or library versions), a set of prioritized context files under .github/copilot, and codebase-scanning instructions for inferring naming conventions, code organization, error handling, logging, and testing patterns. Conditional sections emit quality standards and documentation requirements tailored to the selected focus areas, always instructing Copilot to follow the most consistent existing patterns and never introduce patterns absent from the codebase.
It targets teams and maintainers who want repeatable, standards-aligned AI code generation across a repository. It is a documentation-generation prompt: it reads codebase patterns and emits a Markdown instructions file, with no scripts, credential handling, or destructive operations involved.
A comprehensive copilot-instructions.md file that directs GitHub Copilot to generate code matching your project's standards, architecture, and exact technology versions.
It bases all guidance strictly on actual code patterns in your codebase, instructs version detection so Copilot never uses features beyond the detected versions, and forbids introducing patterns not already present.
No, it is technology-agnostic. A project-type variable supports options like .NET, Java, JavaScript, TypeScript, React, Angular, Python, or multiple, and it adapts accordingly.
Yes. A code-quality focus variable lets you emphasize maintainability, performance, security, accessibility, testability, or all of them, and matching sections are generated conditionally.
It prioritizes files under the .github/copilot directory such as architecture.md, tech-stack.md, coding-standards.md, folder-structure.md, and exemplars.md when they exist.
Quick Setup:
.claude/skills/Repository
github/awesome-copilot