microsoft/a11y-llm-eval

An eval tool to benchmark how well LLMs generate accessible HTML

About the project

A11y LLM Evaluation Harness and Dataset is a test suite for measuring how well language models generate accessible HTML. It runs agent-generated pages in a browser and evaluates them with automated accessibility checks.

2 files for Claude Code and GitHub Copilot: Git Commit, a11y-llm-eval copilot-instructions.md — 533 tokens loaded in every session.

Open on GitHub microsoft.github.io

Instructions

a11y-llm-eval copilot-instructions.md .github/copilot-instructions.md A 533 tok

Agents

Git Commit .github/agents/git-commit.agent.md A 34 tok

These files are microsoft/a11y-llm-eval's own configuration — they tell Claude Code and GitHub Copilot how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.