← ClaudeAtlas

ops-browserlisted

Use when directly operating or verifying a specified page, capturing same-state selected-source/runtime visual and computed evidence, or gathering evidence for an isolated browser-layer failure; require an available browser, verified target, and proven capability, not external-AI orchestration, desktop-client proof, or cross-system diagnosis.
idaibin/skills · ★ 6 · Web & Frontend · score 78
Install: claude install-skill idaibin/skills
# Ops Browser ## Overview Operate browser pages and collect evidence without conflating browser surfaces. The Codex in-app Browser, ChatGPT cloud/agent browser, controlled Chrome, and isolated managed automation have different state, login, download, visibility, and background guarantees. This catalog classifies the host-provided Codex in-app Browser as a non-interrupting host surface; that classification is not an official public browser API guarantee and says nothing about which inspection features it exposes. Select evidence only from capabilities available in the active environment, and leave frontend code changes to `dev-frontend`. ## Workflow 1. Identify the target hostname, path, environment, account/session, and task goal. For an ordinary desktop operation with no requested viewport or matrix, use the package default `1920 x 1080` CSS pixels. Otherwise resolve the viewport set with the precedence and exceptions in `references/usage.md`; do not add a category the user did not request. Verify the effective page viewport rather than treating a successful resize call as proof. 2. Preflight only task-required capabilities and return the Capability Snapshot from `references/browser-operation-protocol.md`. Set unselected availability fields to `unknown` and explain `not assessed: outside selected preflight scope` in `gaps.reason`; expand the matrix only for authenticated, state-changing, transfer, delegated review, or explicitly non-interrupting work. 3. B