Token导航 LogoToken导航

Agent Skill · 研究检索

pdfkit(pdfkit 搜索)

pdfkit 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。

pdfkit 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,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 9,067。

总安装 14,982周安装 637GitHub Stars 535下载 5,249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:pdfkit(pdfkit 搜索)
来源仓库:https://github.com/dpearson2699/swift-ios-skills
仓库路径:skills/pdfkit
安装命令:
npx skills add https://github.com/dpearson2699/swift-ios-skills --skill pdfkit
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/dpearson2699/swift-ios-skills --skill pdfkit

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。

pdfkit 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词或任务场景快速定位候选结果。可通过 npx skills add 命令从指定仓库安装使用。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。

pdfkit 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/pdfkit

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/dpearson2699/swift-ios-skills --skill pdfkit;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

    它需要需要联网,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

pdfkit 怎么安装?

在终端执行 npx skills add https://github.com/dpearson2699/swift-ios-skills --skill pdfkit。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

pdfkit 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。

pdfkit 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

pdfkit 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。

pdfkit 用的人多吗?

公开数据是累计安装 14,982 次,近一周安装 637 次,下载 5,249 次,GitHub 535 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。

pdfkit 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。

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。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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
pdfkit

pdfkit 搜索 · 本页

14,982535需要联网未明确有提醒

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

PDFKit

Display, navigate, search, annotate, and manipulate PDF documents with PDFView, PDFDocument, PDFPage, PDFAnnotation, and PDFSelection. Targets Swift 6.3 / iOS 26+.

Contents

Setup

PDFKit requires no entitlements or Info.plist entries.

import PDFKit

Platform availability: iOS 11+, iPadOS 11+, Mac Catalyst 13.1+, macOS 10.4+, tvOS 11+, visionOS 1.0+.

Displaying PDFs

PDFView is a UIView subclass that renders PDF content, handles zoom, scroll, text selection, and page navigation out of the box.

import PDFKit
import UIKit

class PDFViewController: UIViewController {
    let pdfView = PDFView()

    override func viewDidLoad() {
        super.viewDidLoad()
        pdfView.frame = view.bounds
        pdfView.autoresizingMask = [.flexibleWidth, .flexibleHeight]
        view.addSubview(pdfView)

        pdfView.autoScales = true
        pdfView.displayMode = .singlePageContinuous
        pdfView.displayDirection = .vertical

        if let url = Bundle.main.url(forResource: "sample", withExtension: "pdf") {
            pdfView.document = PDFDocument(url: url)
        }
    }
}

Display Modes

ModeBehavior
.singlePageOne page at a time
.singlePageContinuousPages stacked vertically, scrollable
.twoUpTwo pages side by side
.twoUpContinuousTwo-up with continuous scrolling

Scaling and Appearance

pdfView.autoScales = true
pdfView.minScaleFactor = pdfView.scaleFactorForSizeToFit
pdfView.maxScaleFactor = 4.0

pdfView.displaysPageBreaks = true
pdfView.pageShadowsEnabled = true
pdfView.interpolationQuality = .high

Loading Documents

PDFDocument loads from a URL, Data, or can be created empty.

let fileDoc = PDFDocument(url: fileURL)
let dataDoc = PDFDocument(data: pdfData)
let emptyDoc = PDFDocument()

Password-Protected PDFs

guard let document = PDFDocument(url: url) else { return }
if document.isLocked {
    if !document.unlock(withPassword: userPassword) {
        // Show password prompt
    }
}

Saving and Page Manipulation

document.write(to: outputURL)
document.write(to: outputURL, withOptions: [
    .ownerPasswordOption: "ownerPass", .userPasswordOption: "userPass"
])
let data = document.dataRepresentation()

// Pages (0-based)
let count = document.pageCount
document.insert(PDFPage(), at: count)
document.removePage(at: 2)
document.exchangePage(at: 0, withPageAt: 3)

Page Navigation

PDFView provides built-in navigation with history tracking.

// Go to a specific page
if let page = pdfView.document?.page(at: 5) {
    pdfView.go(to: page)
}

// Sequential navigation
pdfView.goToNextPage(nil)
pdfView.goToPreviousPage(nil)
pdfView.goToFirstPage(nil)
pdfView.goToLastPage(nil)

// Check navigation state
if pdfView.canGoToNextPage { /* ... */ }

// History navigation
if pdfView.canGoBack { pdfView.goBack(nil) }

// Go to a specific point on a page
let destination = PDFDestination(page: page, at: CGPoint(x: 0, y: 500))
pdfView.go(to: destination)

Observing Page Changes

NotificationCenter.default.addObserver(
    self, selector: #selector(pageChanged),
    name: .PDFViewPageChanged, object: pdfView
)

@objc func pageChanged(_ notification: Notification) {
    guard let page = pdfView.currentPage,
          let doc = pdfView.document else { return }
    let index = doc.index(for: page)
    pageLabel.text = "Page \(index + 1) of \(doc.pageCount)"
}

Text Search and Selection

Synchronous Search

let results: [PDFSelection] = document.findString(
    "search term", withOptions: [.caseInsensitive]
)

Asynchronous Search

Use PDFDocumentDelegate for background searches on large documents. Implement didMatchString(_:) to receive each match and documentDidEndDocumentFind(_:) for completion.

Incremental Search and Find Interaction

// Find next match from current selection
let next = document.findString("term", fromSelection: current, withOptions: [.caseInsensitive])

// System find bar (iOS 16+)
pdfView.isFindInteractionEnabled = true

Text Extraction

let fullText = document.string                          // Entire document
let pageText = document.page(at: 0)?.string             // Single page
let attributed = document.page(at: 0)?.attributedString  // With formatting

// Region-based extraction
if let page = document.page(at: 0) {
    let selection = page.selection(for: CGRect(x: 50, y: 50, width: 400, height: 200))
    let text = selection?.string
}

Highlighting Search Results

let results = document.findString("important", withOptions: [.caseInsensitive])
for selection in results { selection.color = .yellow }
pdfView.highlightedSelections = results

if let first = results.first {
    pdfView.setCurrentSelection(first, animate: true)
    pdfView.go(to: first)
}

Annotations

Annotations are created with PDFAnnotation(bounds:forType:withProperties:) and added to a PDFPage.

Highlight Annotation

func addHighlight(to page: PDFPage, selection: PDFSelection) {
    let highlight = PDFAnnotation(
        bounds: selection.bounds(for: page),
        forType: .highlight, withProperties: nil
    )
    highlight.color = UIColor.yellow.withAlphaComponent(0.5)
    page.addAnnotation(highlight)
}

Text Note Annotation

let note = PDFAnnotation(
    bounds: CGRect(x: 100, y: 700, width: 30, height: 30),
    forType: .text, withProperties: nil
)
note.contents = "This is a sticky note."
note.color = .systemYellow
note.iconType = .comment
page.addAnnotation(note)

Free Text Annotation

let freeText = PDFAnnotation(
    bounds: CGRect(x: 50, y: 600, width: 300, height: 40),
    forType: .freeText, withProperties: nil
)
freeText.contents = "Added commentary"
freeText.font = UIFont.systemFont(ofSize: 14)
freeText.fontColor = .darkGray
page.addAnnotation(freeText)

Link Annotation

let link = PDFAnnotation(
    bounds: CGRect(x: 50, y: 500, width: 200, height: 20),
    forType: .link, withProperties: nil
)
link.url = URL(string: "https://example.com")
page.addAnnotation(link)

// Internal page link
link.destination = PDFDestination(page: targetPage, at: .zero)

Removing Annotations

for annotation in page.annotations {
    page.removeAnnotation(annotation)
}

Annotation Subtypes Reference

SubtypeConstantPurpose
Highlight.highlightText markup (yellow highlight)
Underline.underlineText markup (underline)
StrikeOut.strikeOutText markup (strikethrough)
Text.textSticky note icon
FreeText.freeTextInline text block
Ink.inkFreehand drawing paths
Link.linkURL or page destination
Line.lineStraight line with endpoints
Square.squareRectangle shape
Circle.circleEllipse shape
Stamp.stampRubber stamp (Approved, etc.)
Widget.widgetForm element (text field, checkbox)

Thumbnails

PDFThumbnailView

PDFThumbnailView shows a strip of page thumbnails linked to a PDFView.

let thumbnailView = PDFThumbnailView()
thumbnailView.pdfView = pdfView
thumbnailView.thumbnailSize = CGSize(width: 60, height: 80)
thumbnailView.layoutMode = .vertical
thumbnailView.translatesAutoresizingMaskIntoConstraints = false
view.addSubview(thumbnailView)

Generating Thumbnails Programmatically

let thumbnail = page.thumbnail(of: CGSize(width: 120, height: 160), for: .mediaBox)

// All pages
let thumbnails = (0..<document.pageCount).compactMap {
    document.page(at: $0)?.thumbnail(of: CGSize(width: 120, height: 160), for: .mediaBox)
}

SwiftUI Integration

Wrap PDFView in a UIViewRepresentable for SwiftUI.

import SwiftUI
import PDFKit

struct PDFKitView: UIViewRepresentable {
    let document: PDFDocument

    func makeUIView(context: Context) -> PDFView {
        let pdfView = PDFView()
        pdfView.autoScales = true
        pdfView.displayMode = .singlePageContinuous
        pdfView.document = document
        return pdfView
    }

    func updateUIView(_ pdfView: PDFView, context: Context) {
        if pdfView.document !== document {
            pdfView.document = document
        }
    }
}

Usage

struct DocumentScreen: View {
    let url: URL

    var body: some View {
        if let document = PDFDocument(url: url) {
            PDFKitView(document: document)
                .ignoresSafeArea()
        } else {
            ContentUnavailableView("Unable to load PDF", systemImage: "doc.questionmark")
        }
    }
}

For interactive wrappers with page tracking, annotation hit detection, and coordinator patterns, see references/pdfkit-patterns.md.

Page Overlays (iOS 16+)

PDFPageOverlayViewProvider places UIKit views on top of individual pages for interactive controls or custom rendering beyond standard annotations.

class OverlayProvider: NSObject, PDFPageOverlayViewProvider {
    func pdfView(_ view: PDFView, overlayViewFor page: PDFPage) -> UIView? {
        let overlay = UIView()
        // Add custom subviews
        return overlay
    }
}

pdfView.pageOverlayViewProvider = overlayProvider

Common Mistakes

DON'T: Force-unwrap PDFDocument init

PDFDocument(url:) and PDFDocument(data:) are failable initializers.

// WRONG
let document = PDFDocument(url: url)!

// CORRECT
guard let document = PDFDocument(url: url) else { return }

DON'T: Forget autoScales on PDFView

Without autoScales, the PDF renders at its native resolution.

// WRONG
pdfView.document = document

// CORRECT
pdfView.autoScales = true
pdfView.document = document

DON'T: Ignore PDF coordinate system in annotations

PDF page coordinates have origin at the bottom-left with Y increasing upward -- opposite of UIKit.

// WRONG: UIKit coordinates
let bounds = CGRect(x: 50, y: 50, width: 200, height: 30)

// CORRECT: PDF coordinates (origin bottom-left)
let pageBounds = page.bounds(for: .mediaBox)
let pdfY = pageBounds.height - 50 - 30
let bounds = CGRect(x: 50, y: pdfY, width: 200, height: 30)

DON'T: Modify annotations on a background thread

PDFKit classes are not thread-safe.

// WRONG
DispatchQueue.global().async { page.addAnnotation(annotation) }

// CORRECT
DispatchQueue.main.async { page.addAnnotation(annotation) }

DON'T: Compare PDFDocument with == in UIViewRepresentable

PDFDocument is a reference type. Use identity (!==).

// WRONG: Always replaces document
func updateUIView(_ pdfView: PDFView, context: Context) {
    pdfView.document = document
}

// CORRECT
func updateUIView(_ pdfView: PDFView, context: Context) {
    if pdfView.document !== document {
        pdfView.document = document
    }
}

Review Checklist

  • PDFDocument init uses optional binding, not force-unwrap
  • pdfView.autoScales = true set for proper initial display
  • Page indices checked against pageCount before access
  • displayMode and displayDirection configured to match design
  • Annotations use PDF coordinate space (origin bottom-left, Y up)
  • All PDFKit mutations happen on the main thread
  • Password-protected PDFs handled with isLocked / unlock(withPassword:)
  • SwiftUI wrapper uses !== identity check in updateUIView
  • PDFViewPageChanged notification observed for page tracking
  • PDFThumbnailView.pdfView linked to the main PDFView
  • Large-document search uses async beginFindString with delegate
  • Saved documents use write(to:withOptions:) when encryption needed

References

平台分布

Codex

38.61%
按下载量换算2,027

Claude

32.69%
按下载量换算1,716

Cursor

16.73%
按下载量换算878

Gemini CLI

9%
按下载量换算472

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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