Token导航 LogoToken导航

Agent Skill · 研究检索

frappe-impl-website(frappe impl 网站)

frappe-impl-website 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

frappe-impl-website 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,674 个 Skill 里排第 72,323。

总安装 188周安装 8GitHub Stars 79下载 66核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:frappe-impl-website(frappe impl 网站)
来源仓库:https://github.com/openaec-foundation/frappe_claude_skill_package
仓库路径:skills/frappe-impl-website
安装命令:
npx skills add https://github.com/openaec-foundation/frappe_claude_skill_package --skill frappe-impl-website
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/openaec-foundation/frappe_claude_skill_package --skill frappe-impl-website

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

frappe-impl-website 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果时使用。通过 npx skills add 命令从指定 GitHub 仓库安装,需结合原始 README 核验具体用法。

安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写操作。适用于网站内容管理和页面实现场景。

来源仓库内路径:skills/frappe-impl-website

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/openaec-foundation/frappe_claude_skill_package…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

frappe-impl-website 怎么安装?

在终端执行 npx skills add https://github.com/openaec-foundation/frappe_claude_skill_package --skill f…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

frappe-impl-website 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

frappe-impl-website 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

frappe-impl-website 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

frappe-impl-website 用的人多吗?

公开数据是累计安装 188 次,近一周安装 8 次,下载 66 次,GitHub 79 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

frappe-impl-website 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
frappe-impl-website

frappe impl 网站 · 本页

18879操作浏览器未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Frappe Website & Portals — Implementation Workflows

Step-by-step workflows for building websites, portals, and public-facing pages. For hooks syntax see frappe-impl-hooks. For Jinja templating see frappe-impl-jinja.

Version: v14/v15/v16 | Note: v15+ uses Bootstrap 5; v14 uses Bootstrap 4.

Quick Decision: Which Page Type?

WHAT do you need?
├── Static content page (About, Terms)     → Web Page DocType or www/ HTML
├── Data entry by external users           → Web Form
├── List of records visible on website     → has_web_view on DocType
├── Blog / news articles                   → Blog Post + Blog Category
├── Custom app with sidebar/toolbar        → Custom Portal Page (www/)
└── Dynamic route with parameters          → website_route_rules in hooks.py

See references/decision-tree.md for the complete decision tree.

Workflow 1: Create a Portal Page (www/)

Portal pages live in your app's www/ directory. The file name becomes the URL route.

  1. Create myapp/www/custom_page.html:
{% extends "templates/web.html" %}
{% block page_content %}
<h1>{{ title }}</h1>
<div>{{ content }}</div>
{% endblock %}
  1. Create matching controller myapp/www/custom_page.py:
import frappe

def get_context(context):
    context.title = "My Custom Page"
    context.content = "Hello World"
    context.no_cache = 1  # ALWAYS set for dynamic content
  1. Result: page available at /custom_page

File types auto-loaded: .html (template), .py (controller), .css (styles), .js (scripts).

Subdirectory pattern — for nested routes:

myapp/www/
├── services/
│   ├── index.html        → /services
│   ├── index.py
│   ├── consulting.html   → /services/consulting
│   └── consulting.py

Context Variables Reference

KeyTypeEffect
titlestrPage title and browser tab
no_cacheboolDisable page caching
no_headerboolHide the page header
no_breadcrumbsboolRemove breadcrumbs
add_breadcrumbsboolAuto-generate from folder structure
show_sidebarboolDisplay web sidebar
sitemapint0 = exclude from sitemap, 1 = include
metatagsdictSEO meta tags (see Workflow 7)

Rule: ALWAYS set no_cache = 1 for pages with user-specific or frequently changing content.

Workflow 2: Create a Web Form

Web Forms let external users submit data that creates Frappe documents.

  1. Navigate to Web Form list → New Web Form
  2. Set Title, select target DocType, set Route (URL slug)
  3. Add fields — ALWAYS match fieldname to the target DocType field names
  4. Configure access:

- Login Required: uncheck for guest submissions - Allow Edit: let users edit their submissions - Allow Multiple: let users submit more than once

  1. Save and publish

Guest Submissions

ALLOWING guest submissions?
├── YES → Uncheck "Login Required"
│        → Set "Guest Title" for the submission form
│        → ALWAYS add rate limiting in site_config:
│           "rate_limit": {"web_form": "5/hour"}
│        → ALWAYS validate server-side (guests can bypass JS)
└── NO  → Keep "Login Required" checked (default)

Web Form Custom Script (Client)

frappe.web_form.on("after_load", function() {
    // Runs after form loads in browser
});

frappe.web_form.on("before_submit", function() {
    // Validate before submission — return false to cancel
    let val = frappe.web_form.get_value("email");
    if (!val) {
        frappe.throw("Email is required");
        return false;
    }
});

frappe.web_form.on("after_submit", function() {
    // Redirect or show message after success
    window.location.href = "/thank-you";
});

Web Form Custom Script (Server: Python)

In the Web Form document, add a Python script:

def get_context(context):
    # Add custom context variables for the template
    context.categories = frappe.get_all("Category", fields=["name", "title"])

Rule: NEVER trust client-side validation alone for Web Forms. ALWAYS validate in the target DocType's controller or server script.

Workflow 3: Enable has_web_view on a DocType

This makes individual documents accessible as web pages (e.g., /articles/my-article).

  1. Open DocType → check Has Web View and Allow Guest to View
  2. Set the Route field prefix (e.g., articles)
  3. ALWAYS add these fields to the DocType:

- route (Data, hidden) — auto-generated URL slug - published (Check) — controls visibility

  1. Create templates in the DocType directory:

- {doctype_name}.html — single record template - {doctype_name}_row.html — list item template

  1. In hooks.py, register as website generator:
website_generators = ["Article"]
  1. In the controller, implement get_context:
class Article(WebsiteGenerator):
    website = frappe._dict(
        template="templates/generators/article.html",
        condition_field="published",
        page_title_field="title",
    )

    def get_context(self, context):
        context.related = frappe.get_all(
            "Article",
            filters={"published": 1, "name": ("!=", self.name)},
            fields=["title", "route"],
            limit=5,
        )

Rule: ALWAYS include a published check field. NEVER expose unpublished documents to guests.

Workflow 4: Website Route Rules (hooks.py)

Route rules map URL patterns to controllers or pages.

# hooks.py
website_route_rules = [
    # Map parameterized URL to a page
    {"from_route": "/projects/<name>", "to_route": "projects/project"},
    # Map URL prefix to DocType
    {"from_route": "/kb/<path:name>", "to_route": "knowledge-base"},
]

# Redirects (301/304)
website_redirects = [
    {"source": "/old-page", "target": "/new-page"},
    {"source": r"/docs(/.*)?", "target": r"https://docs.example.com\1"},
]

# Homepage for logged-in users (role-based)
role_home_page = {
    "Customer": "orders",
    "Supplier": "rfqs",
}

# Dynamic homepage
get_website_user_home_page = "myapp.utils.get_home_page"

Priority order for homepage: get_website_user_home_page > role_home_page > Portal Settings > Website Settings.

Workflow 5: Blog Setup

  1. Create Blog Category documents (e.g., "News", "Updates")
  2. Create Blog Post documents:

- Select category, write content (Markdown or Rich Text) - Set Published and Published On date - Blog route auto-generates as /blog/{slug}

  1. Configure in Website Settings:

- Set blog title - Enable/disable comments

Rule: ALWAYS set Published On date — posts without a date NEVER appear in RSS feeds.

Workflow 6: Website Theme & Custom CSS

Via Website Theme DocType

  1. Navigate to Website Theme → New
  2. Configure: fonts, colors, navbar style, button radius
  3. Add custom CSS in the Custom CSS field
  4. Set as active theme in Website Settings

Via hooks.py

# Inject CSS/JS on all web pages
website_context = {
    "favicon": "/assets/myapp/images/favicon.png",
}

update_website_context = "myapp.overrides.website_context"

# Override base template
base_template = "myapp/templates/custom_base.html"

Workflow 7: SEO: Meta Tags, Open Graph & Sitemap

In portal pages (frontmatter or context)

def get_context(context):
    context.metatags = {
        "title": "My Page Title",
        "description": "Page description for search engines",
        "image": "/assets/myapp/images/og-image.png",
        "og:type": "website",
        "twitter:card": "summary_large_image",
    }

In Web Page DocType

Set meta fields directly: Meta Title, Meta Description, Meta Image.

Sitemap

  • Frappe auto-generates /sitemap.xml from published Web Pages and has_web_view documents
  • Exclude pages: set sitemap = 0 in context or frontmatter
  • Custom robots.txt: set robots_txt path in site_config.json

Rule: ALWAYS set meta description on public pages. NEVER leave it empty — search engines penalize pages without descriptions.

Workflow 8: Guest Access & Security

# site_config.json — rate limiting
{
    "rate_limit": {
        "web_form": "5/hour",
        "api": "100/hour"
    },
    "allowed_referrers": ["https://mysite.com"],
    "allow_cors": "https://mysite.com"
}

Security rules:

  • ALWAYS enable CSRF protection (default). NEVER set ignore_csrf in production
  • ALWAYS rate-limit guest-accessible endpoints
  • ALWAYS sanitize user input in Web Forms (Frappe does this by default for standard fields)
  • NEVER expose internal DocType names in guest-facing URLs without access control

Anti-Patterns

Anti-PatternCorrect Approach
Hard-coding HTML in get_contextUse Jinja templates with context variables
Skipping no_cache on dynamic pagesALWAYS set no_cache = 1 for user-specific content
Guest Web Form without rate limitingALWAYS configure rate limits for guest forms
Missing published field on has_web_viewALWAYS add published check to prevent data leaks
Using website_route_rules for simple redirectsUse website_redirects instead
Putting business logic in www/ controllersKeep in DocType controllers; www/ is for presentation

See references/anti-patterns.md for expanded anti-patterns with examples.

See Also

  • frappe-impl-hooks — Website hooks in detail
  • frappe-impl-jinja — Jinja templating patterns
  • frappe-impl-controllers — DocType controllers (WebsiteGenerator)
  • frappe-syntax-clientscripts — Client-side API for Web Forms
  • references/generators.md — Portal generators, blog system, custom routing patterns
  • references/workflows.md — Extended workflow walkthroughs
  • references/examples.md — Complete code examples
  • references/decision-tree.md — Full decision tree for page types

平台分布

Codex

31.72%
按下载量换算21

Claude

29.88%
按下载量换算20

Cursor

19.49%
按下载量换算13

Gemini CLI

9.47%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills