JavaScript in-page GUI agent. Control web interfaces with natural language.
Page Agent is a JavaScript library that embeds a natural-language agent inside a web page and lets it operate the page through its text-based DOM. It is used for web copilots, form filling, accessibility, and browser workflows, while the catalogue entries provide agent skills and instructions for using it.
Latest release v1.12.4 — 🌟v1.12.4 · 6 Sept 2026
7 files for Claude Code, Codex and OpenCode: git-cleanup, maintain-model-list, pre-impl-discussion, submit-pr-from-current-changes and 3 more — 1,635 tokens loaded in every session.
AGENTS.md A 1,630 tok CLAUDE.md A 5 tok .agents/skills/git-cleanup/SKILL.md A 39 tok .agents/skills/maintain-model-list/SKILL.md A 53 tok .agents/skills/pre-impl-discussion/SKILL.md A 59 tok .agents/skills/submit-pr-from-current-changes/SKILL.md A 34 tok .agents/skills/update-changelog/SKILL.md A 52 tok These files are alibaba/page-agent's own configuration — they tell Claude Code, Codex and OpenCode 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.