Drive a headless Chromium browser for UI verification and content scraping.
npx skills add https://github.com/mheers/opencode-sandbox --skill agent-browser-mheers
Please help me install this Agent Skill. Skill: agent-browser Source: https://github.com/mheers/opencode-sandbox/tree/main/skills/agent-browser Command: npx skills add https://github.com/mheers/opencode-sandbox --skill agent-browser-mheers
Real-world AI workflows often need to verify, test, or gather data from live web pages. This skill provides a controlled interface to drive a real browser, enabling reproducible interactions in a headless or visible mode.
Open a URL, snapshot the page to identify element refs, then interact with elements using those refs.
High-intent search queries and answers about installing and using this skill.
You can automate browser interactions by driving a real headless Chromium instance using a snapshot-and-ref workflow. You open a URL, take a snapshot to identify stable element refs, then interact with those elements to execute multi-step browser tasks.
A snapshot-and-ref workflow captures the current page state to assign stable element references like @e1 or @e2. You use these refs to target elements for actions such as click and fill, ensuring interactions remain stable across page navigations during UI verification.
Yes, headless Chromium supports local-dev server testing and content scraping. You launch the browser instance, navigate to your target URLs, and use commands like get text and get attr to extract content and verify UI states.
You verify UI states in a headless browser by opening the target URL, taking a snapshot to get element refs, and executing interaction commands. The snapshot captures the current page structure so you can assert content and attributes against expected values.
The best way to drive real browser workflows is using a headless Chromium instance with a structured command sequence. You open a page, snapshot it for element refs, perform actions like click and fill, and close the browser once the workflow is complete.
No external dependencies are required to run the headless Chromium browser automation. The skill operates independently to launch the browser instance and execute the snapshot-and-ref interaction commands for your web testing and scraping tasks.
DaviRain-SuAutomate browser-based QA testing with a headless Chromium instance.
abhishekmmgnExtract text, tables, and metadata from web pages via Bash.
MinhHoangDonoAutomate browser interactions using snapshot-based element references for AI agents.
SoSoBurnerImport Chromium browser cookies into a headless Playwright session.
