Navigate web pages with persistent headless Chromium and capture accessibility trees, logs, and screenshots.
npx skills add https://github.com/Soria-Inc/soria-stack --skill browse-soria-inc
Please help me install this Agent Skill. Skill: browse Source: https://github.com/Soria-Inc/soria-stack/tree/main/browse Command: npx skills add https://github.com/Soria-Inc/soria-stack --skill browse-soria-inc
Provide a fast, persistent headless Chromium instance that lets AI agents and engineers verify UI renders, reproduce bugs with rich evidence, inspect third-party pages before building scrapers, and dogfood interactive user flows without manual browser juggling.
Ask browse to open the target URL, take an interactive snapshot, check console errors, and save an annotated screenshot for evidence.
High-intent search queries and answers about installing and using this skill.
Headless Chromium can capture interactive snapshots that produce accessibility trees with @e refs, annotated screenshots, and unified diffs. You navigate to a target URL, take the snapshot, and generate element-level assertions and visual evidence automatically.
Yes, a persistent server-backed browser maintains shared cookies, tabs, and localStorage for fast follow-up probes. This allows you to resume sessions across multiple runs without re-authenticating or losing page state.
Use a headless Chromium instance to inspect third-party pages before building scrapers. It captures network logs, console output, and CSS inspection data, allowing you to verify page structure and dynamic content requirements safely.
Yes, you can reproduce and file bugs with rich evidence by capturing console and network logs alongside annotated screenshots. This provides a complete profile of page behavior for verifying UI renders and documenting failures.
You can use built-in cookie import utilities to inject authentication data directly into the persistent headless Chromium instance. This enables immediate testing of authenticated user flows and dogfooding staging or production environments.
You should use the safe handoff utility to switch to a headed browser when human interaction is required. This is necessary for workflows that involve manual verification, CAPTCHA solving, or complex visual inspection that automated snapshots cannot fully cover.
lyh2Import and decrypt Chromium cookies into a Playwright headless context.
DariusCorvusDrive automated QA browser tasks with navigation, interaction, and state checks.
24601Control a persistent headless Chromium daemon via a local CLI for browser testing.
zhayujieFetch web page HTML and extract title and readable text.
