core-agent-browser — Claude Code skill: browser automation and setup
This skill guides agents through browser tasks with the agent-browser CLI, from inspecting page controls to filling forms. Use it when browser automation is explicitly requested.
npx skills add actionbook/rust-skills --skill core-agent-browser -a claude-codeWhat core-agent-browser does
core-agent-browser describes how to control a browser from the agent-browser CLI: open a page, inspect interactive controls, fill fields, click buttons, and check what happened. A page snapshot assigns references such as @e1 to elements for later commands.
For Rust and crate research, the listed priority is rust-learner, then the actionbook MCP, with direct agent-browser use reserved for sites without prepared selectors or tasks involving interactive checks, screenshots, or form entry. The skill is marked for internal use and is not intended for direct user invocation. The agent-browser command must be available in the environment.
What it helps with
- inspect a page’s interactive controls
- fill and submit a browser form
- check a page after navigation or DOM changes
- save a screenshot of a page
Example prompts for the agent
- Open
<url>withagent-browser, inspect the controls, and click the button needed for this task. - Fill in the form with
agent-browser, submit it, then inspect the updated page. - Save a full-page screenshot as
path.png.
How to install core-agent-browser
1Skills CLI (Vercel)
npx skills add actionbook/rust-skills --skill core-agent-browser -a claude-code2Manually: put the files in place
~/.claude/skills/core-agent-browser/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/core-agent-browser/
1Skills CLI (Vercel)
npx skills add actionbook/rust-skills --skill core-agent-browser -a codex2Manually: put the files in place
.agents/skills/core-agent-browser/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add actionbook/rust-skills --skill core-agent-browser -a cursor2Manually: put the files in place
.cursor/skills/core-agent-browser/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install actionbook/rust-skills core-agent-browser2Manually: put the files in place
.github/skills/core-agent-browser/SKILL.md1Skills CLI (Vercel)
npx skills add actionbook/rust-skills --skill core-agent-browser -a gemini-cli2Platform CLI
gemini skills install https://github.com/actionbook/rust-skills --path skills/core-agent-browserManually: put the files in place
.devin/skills/core-agent-browser/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add actionbook/rust-skills --skill core-agent-browser -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about core-agent-browser
How do I install core-agent-browser in Claude Code?
Run npx skills add actionbook/rust-skills --skill core-agent-browser -a claude-code. The skill file is at ~/.claude/skills/core-agent-browser/SKILL.md.
What does core-agent-browser do?
It guides browser control through the agent-browser CLI, including page navigation, inspecting controls, form interaction, and screenshots.
When should I use agent-browser instead of actionbook?
For Rust and crate research, the listed order starts with rust-learner and the actionbook MCP. Use agent-browser when a site lacks prepared selectors or the task needs interactive testing, screenshots, or form filling.