Token导航 LogoToken导航

Agent Skill · 研究检索

sharex-feature-specifications(sharex 功能规格)

sharex-feature-specifications 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。

sharex-feature-specifications 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 73,675 个 Skill 里排第 8,680。

总安装 15,849周安装 726GitHub Stars 239下载 5,771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:sharex-feature-specifications(sharex 功能规格)
来源仓库:https://github.com/sharex/xerahs
仓库路径:skills/sharex-feature-specifications
安装命令:
npx skills add https://github.com/sharex/xerahs --skill 'ShareX Feature Specifications'
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/sharex/xerahs --skill 'ShareX Feature Specifications'

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。

sharex-feature-specifications 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、任务场景或来源线索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。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/sharex-feature-specifications

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/sharex/xerahs --skill 'ShareX Feature Specific…;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

sharex-feature-specifications 怎么安装?

在终端执行 npx skills add https://github.com/sharex/xerahs --skill 'ShareX Feature Specifications'。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

sharex-feature-specifications 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。

sharex-feature-specifications 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。

sharex-feature-specifications 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。

sharex-feature-specifications 用的人多吗?

公开数据是累计安装 15,849 次,近一周安装 726 次,下载 5,771 次,GitHub 239 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。

sharex-feature-specifications 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
sharex-feature-specifications

sharex 功能规格 · 本页

15,849239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

Uploader Plugin System Specification

Architecture Overview

Multi-Instance Provider Catalog:

  • Renamed IUploaderPlugin → IUploaderProvider with multi-category support
  • Separated provider (type) from instance (configured occurrence)
  • ProviderCatalog: Static registry for provider types
  • InstanceManager: Singleton for instance lifecycle, persistence, default selection
  • Models: UploaderInstance, InstanceConfiguration with JSON serialization
  • UI: ProviderCatalogViewModel, CategoryViewModel, UploaderInstanceViewModel
  • Full CRUD operations: Add from catalog, duplicate, rename, remove, set default
  • Cross-category support: Same provider (e.g., S3) can serve Image + Text + File

Providers Reference:

  • ImgurProvider: Supports Image + Text categories
  • AmazonS3Provider: Supports Image + Text + File categories

Persistence: %AppData%/XerahS/uploader-instances.json

Dynamic Plugin System

Architecture:

  • Pure Dynamic Loading: No compile-time plugin references in host app.
  • Isolation: PluginLoadContext (AssemblyLoadContext) for each plugin.
  • Shared Dependencies: Framework assemblies (Avalonia, Newtonsoft.Json) shared from host context.
  • Static Lifecycle: Static PluginLoader prevents premature GC of contexts.
  • Dynamic UI: Plugins expose configuration views via IUploaderProvider.GetConfigView.

Components:

  • PluginDiscovery: Scans Plugins/ folder for plugin.json manifests.
  • PluginLoader: Loads assemblies and instantiates providers.
  • ProviderCatalog: Central registry for both built-in and dynamic providers.

Plugin Examples:

  • ShareX.Imgur.Plugin: Image uploading with OAuth2.
  • ShareX.AmazonS3.Plugin: S3 bucket uploads (Image/Text/File).

File-Type Routing (Planned)

Goals:

  • Deterministic routing based on file extension
  • Conflict prevention (no overlapping file types per category)
  • "All File Types" as exclusive option
  • UI showing available/blocked file types

Data model additions:

  • FileTypeScope class (AllFileTypes flag + FileExtensions list)
  • UploaderInstance.FileTypeRouting property
  • IUploaderProvider.GetSupportedFileTypes() method

Routing logic:

1. Exact extension match (e.g., .png → Imgur)
2. "All File Types" fallback
3. No match → error/notification

UI features:

  • File type selector with conflict detection
  • Disabled checkboxes for already-assigned types
  • Tooltip showing which instance blocks a type
  • Real-time validation

Implementation phases:

  1. Data model extensions
  2. Routing engine + validation
  3. Provider metadata
  4. UI (file type selector, conflict warnings)
  5. Upload workflow integration

Full Automation Workflow (Path B - Planned)

Goal: Complete end-to-end automation matching ShareX feature parity

Components needed:

  1. Task System (~600 LOC)

- WorkflowTask.cs - Async automation engine with lifecycle management - TaskInfo.cs - Task metadata and state - TaskManager.cs - Queue orchestration with concurrency control - TaskStatus.cs - Status enums and job types

  1. Workflow Configuration (~200 LOC)

- AfterCaptureTasks flags: AddImageEffects, AnnotateImage, SaveImageToFile, UploadImageToHost, etc. - AfterUploadTasks flags: CopyURLToClipboard, OpenURL, ShortenURL, ShowQRCode, ShowNotification - UploadInfoParser - Template-based clipboard format (%url%, %filename%, %time%, etc.)

  1. Upload Integration (~150 LOC)

- UploadResult model with URL/ThumbnailURL/DeletionURL/ShortenedURL - UploaderErrorManager for error handling - Progress tracking events - Retry logic with secondary uploaders

  1. After-Upload Automation (~100 LOC)

- URL shortening integration - Clipboard copy with custom formats - Browser opening - QR code generation - Notification system

  1. UI Integration (~150 LOC)

- Progress indicator during upload - Toast notifications - Upload history panel - Task list view

Implementation phases:

  1. Core task system (WorkflowTask, TaskManager)
  2. Upload workflow integration
  3. After-upload automation (clipboard, notifications)
  4. Hotkey → full automation wiring
  5. Progress UI and history

Total effort: ~1,200 LOC, 8-12 hours

Blockers:

  • Annotation editor is blocking (modal). Need separate workflow path for "quick capture+upload" vs "editor mode"
  • Need async/await patterns throughout (ShareX uses threads)
  • Upload providers need UploadAsync method signatures

Path A (minimal) delivered:

  • Basic WorkflowTask structure
  • Upload provider wiring
  • Simple "Copy URL after upload" to clipboard
  • Hotkey → Capture → Upload → URL workflow

Path B additions:

  • Full task queue with concurrency limits
  • Comprehensive AfterCapture/AfterUpload pipelines
  • Progress tracking with percentage/speed/ETA
  • Notification system with customizable formats
  • Upload retry logic with fallback uploaders
  • Task history persistence
  • UI for task monitoring and management

Annotation Subsystem Specification

Phase 1: Core Annotation Models (Completed)

Project: XerahS.Annotations Files created: 10 files (~800 LOC)

Annotation types implemented:

  • EditorTool enum: Select, Rectangle, Ellipse, Arrow, Line, Text, Number, Spotlight, Crop, Pen
  • Annotation abstract base class with rendering, hit-testing, bounds calculation
  • Concrete types: RectangleAnnotation, EllipseAnnotation, LineAnnotation, ArrowAnnotation, TextAnnotation, NumberAnnotation (auto-increment), SpotlightAnnotation (focus effect), CropAnnotation (resize handles)

Technical features:

  • Avalonia DrawingContext rendering
  • Configurable hit testing (tolerance-based)
  • Color/stroke width support
  • Z-index for layering
  • Manual distance calculations (Avalonia API compatibility)
  • FormattedText for text rendering with bold/italic support

Phase 2: Canvas Control and Full Feature Set

Implement the annotation subsystem for XerahS, replacing WinForms and System.Drawing with Avalonia and Skia. All features listed in the ShapeType enum must be available.

Design Core Abstractions
  • Define a BaseShape in XerahS with properties for position, size, colour, border thickness and hit-testing.
  • Create Avalonia equivalents for all ShapeType values (RegionRectangle, RegionEllipse, RegionFreehand, DrawingRectangle, DrawingEllipse, DrawingFreehand, DrawingFreehandArrow, DrawingLine, DrawingArrow, DrawingTextOutline, DrawingTextBackground, DrawingSpeechBalloon, DrawingStep, DrawingMagnify, DrawingImage, DrawingImageScreen, DrawingSticker, DrawingCursor, DrawingSmartEraser, EffectBlur, EffectPixelate, EffectHighlight, ToolSpotlight, ToolCrop, ToolCutOut). Each must subclass BaseShape and implement custom rendering logic.
  • Expose shape-specific properties such as arrow-head direction, blur radius, pixel size and highlight colour with appropriate defaults.
Rendering and Effects
  • Replace GDI+ drawing with Avalonia's DrawingContext and Skia. Implement vector drawing for shapes and text. For effects (blur, pixelate, highlight), use Skia image filters or custom pixel shaders to apply the effect only within the shape bounds.
  • Implement the spotlight tool by darkening the canvas outside the selected rectangle and optionally applying blur/ellipse.
  • Implement magnify drawing by rendering a zoomed portion of the underlying bitmap inside a circle/rectangle shape.
Interaction
  • Create a RegionCaptureView control that captures mouse/touch events, draws handles for resizing/moving shapes and supports multi-shape editing.
  • Implement keyboard shortcuts to switch between tools (e.g., arrow key, text tool, effect tool) and to cancel or confirm actions.
  • Provide a shape toolbar (in XAML) to select drawing, effect and tool types; include controls for changing colours, thicknesses, font and effect parameters.
Shape Management
  • Port ShapeManager to manage a collection of shapes, ordering, selection and removal. Provide undo/redo functionality and support copy/paste of shapes.
  • Persist user preferences (colours, sizes, arrow head direction, blur radius, pixel size etc.) in AnnotationOptions analogues and load/save them on application start.
Image Insertion and Sticker Support
  • Implement file picker integration for inserting external images (DrawingImage and DrawingImageScreen).
  • Provide a sticker palette and cursor stamp support; allow users to import custom stickers.
Smart Eraser
  • Analyse SmartEraserDrawingShape and implement a mask-based eraser that restores the original screenshot pixels under the eraser path.
Crop and Cut-Out Tools
  • Implement crop and cut-out tools that modify the canvas bitmap and adjust existing shapes accordingly.
  • Ensure non-destructive editing by allowing the user to revert or adjust crop boundaries.
Region Capture Integration
  • Replace RegionCaptureForm with a new RegionCaptureWindow built in Avalonia. Handle monitor selection, last region recall, screen colour picker, ruler and other capture modes defined in RegionCaptureMode and RegionCaptureAction.
  • Provide a seamless workflow from capture → annotate → upload, integrating with existing task runners.
Cross-Platform Considerations
  • Avoid any reference to System.Drawing or WinForms; rely solely on Avalonia and Skia for drawing.
  • Use platform abstraction services (e.g., clipboard, file dialogs) via the core platform layer defined in AGENTS.md.
  • Ensure equal behaviour on Windows, macOS and Linux; implement stubs or fallbacks where OS-specific features cannot be supported.
Testing
  • Develop unit and UI tests for each shape type and effect to verify correct rendering, hit-testing, configuration persistence and behaviour in undo/redo operations.
  • Provide manual test scripts to validate annotation workflows across supported platforms.

平台分布

Codex

35.21%
按下载量换算2,032

Claude

31.65%
按下载量换算1,827

Cursor

20.83%
按下载量换算1,202

Gemini CLI

9.24%
按下载量换算533

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