ワンクリックで
browser
Agent can open a page in the browser runtime, inspect the current page, and return the live state.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Agent can open a page in the browser runtime, inspect the current page, and return the live state.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | browser |
| description | Agent can open a page in the browser runtime, inspect the current page, and return the live state. |
| enabled | true |
| skill_class | system |
| permission_label | Browser runtime |
| execution_mode | live |
| action_class | read |
| connector_scopes | ["browser"] |
| trigger_terms | ["browser","browse","open site","open website","navigate","inspect page","open http","open https"] |
| allowed_runtime_modes | ["hosted_secure","local_secure","privileged_device"] |
| requires_approval | false |
| execution_adapter | browser |
Use the browser runtime to open a page, observe the current state, and return a grounded summary.
Template for a workspace-local business skill backed by a handler or custom adapter.
Agent can check live workspace inventory for stock, fitment, availability, and pricing.
Agent can search the web using the web_search tool. Use when the user asks about current events or facts.
Agent can run Python and shell code, show output, handle errors, and iterate.
Agent can read, write, list, and delete files. Always confirm before deleting.
Agent can save and recall facts about the user. Use memory_search before answering personal questions.