Skip to content

AGENTS.md

Latest commit

 

History

History
25 lines (17 loc) · 1.54 KB

AGENTS.md

File metadata and controls

25 lines (17 loc) · 1.54 KB

Agent Instructions for Daily DAP

This project uses AI agents to maintain code quality and accessibility standards. For detailed agent rules and guidelines, see:

.kittify/AGENTS.md

Quick Reference for AI Agents

When working on this project, always:

  1. Follow Path Reference Rules: Use absolute or project-relative paths (e.g., src/scanners/lighthouse-runner.js)
  2. Use UTF-8 Encoding Only: Never use smart quotes, em dashes, or Windows-1252 characters
  3. Maintain Accessibility: This project benchmarks accessibility - ensure all generated HTML follows WCAG 2.2 AA standards
  4. Preserve Security: Always use escapeHtml() when rendering user-controlled content in HTML
  5. Follow Testing Practices: Run npm test before committing changes
  6. Disclose AI Usage: When you contribute to this project, update the ## AI Disclosure section in README.md to identify the LLM used, the version (if known), and the nature of your contribution (e.g., code implementation, documentation, planning). Do not list tools that were not used.

Project-Specific Context

  • Purpose: Daily benchmarking of U.S. government website quality using Lighthouse and ScanGov
  • Key Components: Accessibility scanning, impact estimation, HTML report generation
  • Testing: Node.js built-in test runner with unit, contract, and integration tests
  • Build: Standard npm scripts, compatible with Node.js >= 22.19

For complete accessibility guidelines specific to this project, see ACCESSIBILITY.md.