This directory contains configuration files for GitHub Copilot coding agent to work effectively with the Schichtplan codebase.
Main instructions file that Copilot reads to understand:
- Project structure and architecture
- Development workflows and commands
- Coding standards and patterns
- Critical constraints and gotchas
- Debugging procedures
This file is comprehensive (329 lines) and covers all aspects of the project.
Automated setup steps that Copilot coding agent uses to:
- Set up Python virtual environment
- Install dependencies (Python + Bun)
- Initialize database
- Run quality checks (linting, type checking)
- Execute test suites
- Build verification
Copilot will run these steps when starting work on the repository to ensure proper environment configuration.
Path-specific instructions that apply when working with particular areas of the codebase:
-
scheduler.instructions.md- Guidelines for backend scheduler module- Interval-based coverage logic
- Keyholder constraints
- Performance patterns
- Debugging resources
-
frontend-components.instructions.md- Frontend component development- Design system patterns
- Type safety rules
- API integration
- AI component guidelines
-
database-migrations.instructions.md- Database migration best practices- Migration workflows
- Common patterns
- Testing procedures
- Emergency handling
Custom agent configurations for specialized tasks:
mcp-integration.agent.md- MCP and AI integration specialist- MCP tool development
- Conversational AI patterns
- Streaming responses
- Background tasks
When GitHub Copilot coding agent works on this repository:
- Repository-wide context: Reads
copilot-instructions.mdfor general guidance - Setup automation: Executes steps from
copilot-setup-steps.yamlto configure environment - Path-specific guidance: When working in specific directories, applies relevant instructions from
instructions/ - Specialized tasks: Can invoke custom agents from
agents/for domain-specific work
This configuration follows GitHub's best practices for Copilot coding agent:
✅ Clear, concise repository-wide instructions
✅ Automated environment setup steps
✅ Path-specific instructions for specialized areas
✅ Custom agents for domain expertise
✅ Comprehensive documentation references
✅ Common pitfalls and debugging guidance
copilot-instructions.md:
- When project structure changes significantly
- When new major features are added
- When development workflows change
- When new constraints or gotchas are discovered
copilot-setup-steps.yaml:
- When dependencies are added/removed
- When build process changes
- When new quality checks are introduced
- When database setup procedures change
Path-specific instructions:
- When module architecture evolves
- When new patterns are established
- When common mistakes are identified
- When debugging procedures change
Agent configurations:
- When new specialized workflows emerge
- When integration patterns change
- When new tools are added
When making significant changes:
- Update relevant instruction files
- Test instructions with Copilot coding agent
- Ensure consistency across all files
- Update cross-references between files
- Verify YAML syntax is valid
- Document the changes
To verify these instructions work well:
-
Assign a well-scoped issue to Copilot:
Issue: Add a new field to the Employee model Acceptance criteria: - Add 'department' field to Employee model - Create and apply migration - Update API endpoints - Add tests -
Review Copilot's approach:
- Does it follow the documented patterns?
- Does it use the correct tools and commands?
- Does it avoid documented pitfalls?
-
Iterate on instructions:
- If Copilot makes mistakes, clarify instructions
- Add new sections for recurring issues
- Remove outdated guidance
- GitHub Copilot Documentation
- Best Practices for Copilot Coding Agent
- Piloting Copilot in Organizations
- Custom Instructions Guide
If you notice:
- Copilot consistently misunderstanding something
- Instructions that are unclear or contradictory
- Missing guidance for common tasks
- Outdated information
Please update the relevant files or open an issue to discuss improvements.
Last Updated: January 10, 2025
Maintained By: Project maintainers