Automates browser interactions to test web UIs, fill forms, and capture screenshots.
npx skills add https://github.com/Thigh-Food/emdash --skill agent-browser-thigh-food
Please help me install this Agent Skill. Skill: agent-browser Source: https://github.com/Thigh-Food/emdash/tree/main/skills/agent-browser Command: npx skills add https://github.com/Thigh-Food/emdash --skill agent-browser-thigh-food
@e1, @e2) and use them to click, fill, type, select, and check elements deterministically. - Screenshots & Inspection: Capture viewport or full-page screenshots, read text/HTML/values from elements, and check visibility, enabled, or checked states. - Sessions & Debugging: Persist cookies and storage across commands with named sessions, view console logs and page errors, and run headed mode for visual debugging. - Use Case: After modifying a form in your Astro/Cloudflare CMS admin panel, open the page, snapshot the interactive elements, fill the title and content fields, click save, and screenshot the result to confirm the change works. ## Quick Start Use the agent-browser skill to open http://localhost:4321, take an interactive snapshot, and verify the page renders correctly.High-intent search queries and answers about installing and using this skill.
Use the agent-browser CLI to open a URL, run snapshot with the -i flag to get interactive element refs, then interact using commands like click, fill, and type with those refs. Finish with a screenshot to verify the result visually.
Open the page with agent-browser open, take an interactive snapshot to find field refs, then use fill @ref "value" for each input and click the submit button ref. Use find label or find placeholder to target fields without refs.
Yes, named sessions keep cookies and storage between commands. Pass --session mytest to each command or set the AGENT_BROWSER_SESSION environment variable to reuse authenticated state across interactions.
Element refs change between snapshots, so always take a fresh snapshot before interacting. Use the -i flag for cleaner interactive-only output, or use find role/label commands to target elements by accessible name instead of refs.
Run with --headed to watch the browser visually, then use agent-browser console and agent-browser errors to inspect logs and page errors. You can also highlight an element or run eval with JavaScript to inspect page state.
ClaudfatherAutomates headless-browser QA testing, visual crawls, and performance benchmarking against running web applications.
TeamSloppyAutomate UI visual verification with screenshot capture and visual diff comparisons.
SilentFlowerExecutes cross-layer end-to-end verification scenarios across UI, API, and database layers.
ag-gridWrites and runs Playwright e2e tests for AG Grid documentation examples.