browser-automation
Use Home23's live browser controller for screenshots, navigation checks, and page extraction.
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
Use Home23's live browser controller for screenshots, navigation checks, and page extraction.
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
Iteratively improve a weak skill through score, tweak, and retest loops. Run against a target skill, define failure mode and rubric, execute the loop, and get a scored report with recommendations.
Research X/Twitter discourse with live search, profile reads, thread follow-up, and watchlists when the user wants sourced, current sentiment or expert takes without posting.
Operate on X/Twitter through the canonical Home23 X skill. Supports official API-backed read/search/post/reply/media/delete plus bird-backed timeline/mentions fallback.
Let a Home23 companion sing in first person by turning persona context into a repeatable vocal identity and song prompt.
Generate an original song, instrumental, or cover track when the ask is fundamentally "make music from this idea".
Turn a taste brief into a custom multi-track playlist plan, then generate the tracks and optional cover art.
| id | browser-automation |
| name | Browser Automation |
| version | 1.0.0 |
| layer | skill |
| runtime | nodejs |
| author | home23 |
| description | Use Home23's live browser controller for screenshots, navigation checks, and page extraction. |
| category | browser |
| keywords | ["browser","screenshot","page","rendered","extract","navigate"] |
| triggers | ["take a screenshot","extract the page text","check if this page loads","inspect the rendered page"] |
| capabilities | [{"navigate":"Verify a page loads and return title plus URL"},{"extract":"Extract visible text from a page or selector"},{"screenshot":"Save a screenshot using the live browser controller"}] |
Use this skill when a live browser pass is a better fit than plain HTTP fetches.
Use browser-automation for:
Input:
{
"url": "https://example.com",
"waitMs": 3000
}
Input:
{
"url": "https://example.com",
"selector": "main",
"waitMs": 3000
}
Input:
{
"url": "https://example.com",
"waitMs": 3000
}
waitMs.web_browse may be cheaper.