# Kuaishou Search

> Extract structured public Kuaishou video search results from the current browser page.

- Skill: `2233admin/kuaishou-search` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add 2233admin/kuaishou-search`
- Raw SKILL.md: https://api.skillmd.com/api/skills/2233admin/kuaishou-search/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: 2233admin (https://skillmd.com/u/2233admin)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/2233admin/kuaishou-search

---


# Kuaishou — Video Search

> Search keyword → bounded structured video results

## Prerequisites

- Browser Act is available.
- The target browser can reach `kuaishou.com`.
- The current session may need human login or verification.

## Execution

Navigate to:

```text
https://www.kuaishou.com/search/video?searchKey={query}
```

Wait for the page to settle, then run:

```bash
python scripts/extract-search.py --max-results 10
```

The result contains the video URL, caption, author, cover, playable media
URL when exposed by the page, publication timestamp, tags, and bounded
engagement statistics.

## Operational boundary

This pack reads the public search state already present in the browser. It
never automates login, captcha solving, or anti-bot bypass. Login, verification,
regional restrictions, and blocked responses are human-handled conditions.

## Limitations

The manifest extracts the initial search result state only. Cursor-based
follow-up requests and comment collection are intentionally out of scope for
this pack; they require a separate pagination contract and should not be
silently represented as complete results.