DOC: Add comprehensive .github/copilot-instructions.md for coding agent onboarding#247
DOC: Add comprehensive .github/copilot-instructions.md for coding agent onboarding#247
Conversation
Co-authored-by: bewithgaurav <8655500+bewithgaurav@users.noreply.github.com>
Review: Minor fixes needed before mergeGreat work on this comprehensive Copilot instructions file! The content is well-structured and will significantly help AI agents work with this repository. I found a few minor inaccuracies that should be fixed: Issues to Fix1. Incorrect file name reference (Line 87)- ├── logging_config.py # Logging configuration
+ ├── logging.py # Logging configurationThe actual file is 2. Incorrect module reference (Line 82)- ├── ddbc_bindings.py # Native extension loader with architecture detection
+ ├── ddbc_bindings.py # Native extension bindings (auto-generated stub)The architecture detection and loading logic is actually in 3. Missing files from architectureConsider adding these important files to the Core Components section:
After FixesOnce these are addressed:
The content is otherwise accurate and comprehensive - good coverage of build system, CI/CD, platform specifics, and contributing guidelines! 🎉 |
|
this PR stands outdated as we'll now be adding new revamped copilot instructions on #406 as mentioned |
This PR adds a comprehensive
.github/copilot-instructions.mdfile to onboard coding agents to the mssql-python repository efficiently. The instructions are designed to significantly reduce the likelihood of failed PRs, build errors, and exploration time for future coding agents.What's Included
The copilot instructions provide detailed information about:
Key Benefits
Reduces PR Rejection Risk: Documents the complex multi-platform build system, required dependencies, and validation processes that agents need to understand to create working code changes.
Minimizes Build Failures: Provides exact command sequences, prerequisites, and expected failure modes (like missing ODBC headers on Linux) with clear explanations.
Accelerates Development: Eliminates the need for agents to discover repository structure, build processes, and testing requirements through trial and error.
The instructions follow the required format (under 2 pages, non-task-specific) and provide comprehensive coverage of the repository's unique characteristics, including its pybind11/CMake native extension build system, platform-specific library distribution, and sophisticated CI/CD matrix across multiple operating systems and architectures.
Future coding agents can now reference these instructions to work effectively with the repository immediately, rather than spending time exploring and potentially making incorrect assumptions about the build and test processes.
💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.