Welcome, AI Agent! You are assisting with the AntV S2 repository. S2 is a high-performance data-driven multi-dimensional analysis table component. Please strictly follow the rules and conventions defined in this file when generating code, writing documentation, or creating pull requests.
When creating or updating documentation (especially in s2-site/docs/), you MUST ALWAYS update both the Chinese (.zh.md) and English (.en.md) versions simultaneously.
Never leave one language out of sync. Ensure proper technical translation that matches the project's tone.
When you are instructed to create a Pull Request:
- You MUST read and strictly follow the template located at
.github/PULL_REQUEST_TEMPLATE.md. - Ensure all sections (type, description, screenshots if applicable, and self-checks) are properly filled out.
- For scheduled automated tasks, use a structured summary detailing what was checked and updated.
We maintain specific AI agent skills and standard operating procedures (SOPs) within this repository.
If an objective relates to predefined skills, you MUST first route to and read the instructions inside the .agent/skills/ directory before proceeding.
Do NOT use Test-Driven Development (TDD). When writing or modifying core logic (e.g., in packages/s2-core/):
- First: Focus on implementing the clean, functional code logic and seek human validation. Do not generate tests immediately to save context and speed up iteration.
- Second: ONLY after the functional code logic has been validated and approved as correct by the human user, you must then write or update corresponding unit tests.
- Hard Rule: The final unit test coverage for your new or modified code must be at least 80%.
If you add or modify a configuration option or API in the core engine, you MUST provide a corresponding implementation example in the Playgrounds (React/Vue) for developers to test and verify the behavior.