← 返回 Skills 市场
259
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install scrapefun
功能描述
Use this skill when OpenClaw needs to operate a scrapefun server through the dedicated OpenClaw-facing APIs. Prefer /api/openclaw/* query and download endpoi...
安全使用建议
This skill appears to do what it claims: it instructs the agent to interact with a Scrapefun/OpenClaw server using an access key. Before installing, confirm the following: (1) Clarify whether OPENCLAW_ACCESS_KEY must be supplied as an environment variable (registry metadata currently omits it) and ensure you only provide a key scoped with minimal necessary permissions (library_query, download_submit, etc.). (2) The skill targets local/hosted APIs (references show http://localhost:4000) — verify the base URL and that you trust the target server. (3) Because source/homepage is unknown, monitor the first uses (network logs or a proxy) to confirm it only calls the documented endpoints and does not send data elsewhere. (4) Only grant the access key to agents or environments you trust and prefer keys with narrowly scoped permissions and an expiration where possible.
功能分析
Type: OpenClaw Skill
Name: scrapefun
Version: 0.0.2
The skill bundle provides a structured interface for managing a media server (scrapefun) via OpenClaw-specific API endpoints. It includes security-conscious instructions for the agent, such as mandatory permission checks before API calls and explicit prohibitions against accessing sensitive internal routes like `/api/openclaw/bootstrap/status` or job management endpoints. No evidence of data exfiltration, malicious execution, or prompt injection was found; the code and instructions (SKILL.md, references/api.md) are strictly aligned with the stated purpose of media library orchestration.
能力评估
Purpose & Capability
The skill's name/description match the endpoints, required header, and operations documented in SKILL.md and references. The declared primary credential (OPENCLAW_ACCESS_KEY) is appropriate for calling OpenClaw-facing APIs. Minor inconsistency: registry metadata lists no required env vars while the skill metadata declares a primaryEnv (OPENCLAW_ACCESS_KEY). This is likely a bookkeeping mismatch rather than functional misalignment, but it should be clarified.
Instruction Scope
SKILL.md limits runtime actions to calling /api/openclaw/* endpoints (with documented narrow fallbacks) and describes clear permission checks and error handling. It does not instruct the agent to read local files, other env vars, or exfiltrate data to arbitrary external endpoints. The references file sets a base URL of http://localhost:4000 (i.e., calls target the local server), which is consistent with operating a local scrapefun instance.
Install Mechanism
Instruction-only skill with no install spec and no bundled code — nothing is written to disk or downloaded by the skill installer. This is the lowest install risk.
Credentials
The single credential in use (OPENCLAW_ACCESS_KEY) is proportional to the task. However, the registry record's 'required env vars' field is empty while SKILL.md/metadata reference a primaryEnv — this should be reconciled so users know which env var to provide. The skill explicitly forbids using other API keys (e.g., OPENCLAW_SCRAPEFUN_API_KEY), which is a helpful constraint.
Persistence & Privilege
always:false and normal autonomous invocation are set (agent may call the skill autonomously, which is standard). The skill does not request persistent system-level privileges or modify other skills' configs.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install scrapefun - 安装完成后，直接呼叫该 Skill 的名称或使用
/scrapefun触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v0.0.2
**Summary: Major authentication and endpoint policy update. Now strictly favors OpenClaw-facing APIs; legacy fallbacks tightly restricted.**
- Prefers dedicated `/api/openclaw/*` endpoints for all main operations (listing, download, scan, media state).
- Only uses generic scrapefun APIs in rare, clearly listed fallback cases—never as silent replacements.
- Explicitly details required OpenClaw permissions for each endpoint; halts if a permission or endpoint is missing.
- Strictly prohibits previously allowed legacy endpoints for media browsing, download, and status checks.
- Response style and examples rewritten to highlight new API order, fallback restrictions, and blockers.
- Updates overview, workflow, and examples for clarity on allowed/disallowed behaviors.
v0.0.1
initial upload
元数据
常见问题
ScrapeFun 是什么？
Use this skill when OpenClaw needs to operate a scrapefun server through the dedicated OpenClaw-facing APIs. Prefer /api/openclaw/* query and download endpoi... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 259 次。
如何安装 ScrapeFun？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install scrapefun」即可一键安装，无需额外配置。
ScrapeFun 是免费的吗？
是的，ScrapeFun 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
ScrapeFun 支持哪些平台？
ScrapeFun 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 ScrapeFun？
由 Haowei Li（@haoweili97）开发并维护，当前版本 v0.0.2。
推荐 Skills