Token导航 LogoToken导航

Agent Skill · 研究检索

linkify(linkify 搜索)

linkify 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。

linkify 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,权限为调用外部服务;许可证为 MIT,暂无第三方扫描结果,按安装量在研究检索类的 73,675 个 Skill 里排第 49,367。

总安装 658周安装 8GitHub Stars 公开资料未说明下载 65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
MIT

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:linkify(linkify 搜索)
来源仓库:https://github.com/hopeoverture/worldbuilding-system
仓库路径:skills/linkify
安装命令:
npx skills add hopeoverture/worldbuilding-system --skill "linkify"
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。

AgentSkills.tonpx skills
npx skills add hopeoverture/worldbuilding-system --skill "linkify"

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。

linkify 用于信息查找与筛选,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位相关内容时使用。支持基于关键词或任务场景的信息检索和结果过滤。

通过 npx skills add hopeoverture/worldbuilding-system --skill "linkify" 安装使用。安装前建议核实维护状态,注意可能触发联网或文件操作。linkify 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/linkify

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 hopeoverture/worldbuilding-system --skill "linkify";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

    它需要调用外部服务,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

linkify 怎么安装?

在终端执行 npx skills add hopeoverture/worldbuilding-system --skill "linkify"。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

linkify 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。

linkify 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

linkify 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

linkify 用的人多吗?

公开数据是累计安装 658 次,近一周安装 8 次,下载 65 次。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。

linkify 安全吗?

没有查到第三方扫描结果,安全性无法从公开资料判断,安装前请自行审查源码、依赖和命令行为。

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。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 没有第三方扫描结果,安全性只能靠自己读源码判断。
  • 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

  • 要求有安全扫描结论才肯安装的团队。
  • 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
linkify

linkify 搜索 · 本页

658—调用外部服务MIT未收录

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

Linkify Entity

Linkify: $ARGUMENTS

Overview

This skill scans an entity file and automatically wraps any mentions of existing entities in [[wikilink]] syntax. It's a focused tool that only adds links to entities that already exist—it does NOT create new entities.

Quick Start

# Linkify a specific entity
/linkify "Aldersgate"

# Preview changes without modifying
/linkify "Aldersgate" --dry-run

# Linkify by file path
/linkify Worlds/Eldermyr/Settlements/Aldersgate.md

# BULK: Linkify all entities in a world
/linkify --world Eldermyr

# BULK: Linkify only Characters in a world
/linkify --world Eldermyr --category Characters

# BULK: Preview bulk changes
/linkify --world Eldermyr --dry-run

Instructions

Step 1: Parse Arguments

Required:

  • Entity path or name

Optional Flags:

FlagPurposeDefault
--dry-runShow what would be linked without making changesfalse
--all-worldsSearch all worlds for entities, not just the source entity's worldfalse
--case-sensitiveRequire exact case matchingfalse
--world [name]BULK MODE: Process all entities in the specified world-
--category [type]With --world: only process entities in this category (Characters, Settlements, etc.)all

Step 2: Locate Entity & Determine World

  1. If path provided (contains / or .md):

- Read the file directly - Extract world name from path: Worlds/[World Name]/...

  1. If name provided:

- Search Worlds/ directories for matching entity - Try exact filename match first - Then fuzzy match on filename and YAML name: field - If multiple matches, list them and ask user to clarify

  1. If not found:

- List similar entities and ask for clarification

Step 3: Build Entity Index

Create a complete index of existing entities:

  1. Scan all files in Worlds/[World Name]/ recursively (or all worlds if --all-worlds)
  2. For each entity file, extract:

- Filename (without.md extension) → primary name - YAML name: field → canonical name - YAML aliases: array → alternative names

  1. Build lookup dictionary (case-insensitive by default):
{
  "Lord Varic Valdren": { path: "Characters/Lord Varic Valdren.md", canonical: "Lord Varic Valdren" },
  "Varic": { path: "Characters/Lord Varic Valdren.md", canonical: "Lord Varic Valdren" },
  "The Merchant Lord": { path: "Characters/Lord Varic Valdren.md", canonical: "Lord Varic Valdren" },
  ...
}

Important: Sort the lookup dictionary by name length (longest first) to ensure longer matches are found before shorter partial matches.

Step 4: Identify Content to Scan

Read the source entity file and identify sections to scan:

INCLUDE these sections:

  • Overview
  • Geography
  • History
  • Demographics
  • Government & Politics
  • Economy
  • Defense & Military
  • Notable Locations
  • Key Figures
  • Secrets
  • Plot Hooks
  • Description
  • Any custom content sections

EXCLUDE these sections:

  • YAML frontmatter (between --- markers)
  • Image Prompts section and everything after it
  • Connections section (already has wikilinks)
  • Code blocks (between ``` markers)
  • Existing wikilinks [[...]]
  • Block quotes used for image prompts (lines starting with > after **Prompt:**)

Step 5: Find Matches

For each entity name/alias in the index (longest first):

  1. Skip if entity is the source file itself (don't self-link)
  2. Search for the name in content:

- Use word-boundary matching to avoid partial matches within words - Match: "Lord Varic Valdren" in "talked to Lord Varic Valdren about" - Skip: "Lord Varic Valdren" in "[[Lord Varic Valdren]]" (already linked) - Skip: "Stone" in "Aldric Stone" if "Aldric Stone" is a separate entity

  1. Record each match:

- Line number - Exact text matched - Canonical entity name (for the wikilink) - Surrounding context (5 words before/after)

  1. Handle partial name matches:

- If "Lord Varic Valdren" exists but text says "Lord Varic", offer to link as [[Lord Varic Valdren|Lord Varic]] - If "House Valdren" exists but text says "the Valdren house", consider display text

Step 6: Present Findings

=== LINKIFY ANALYSIS: [Entity Name] ===

World: [World Name]
Source: [path/to/entity.md]
Entities Indexed: X

---

MATCHES FOUND: Y

| # | Text Found | Line | Links To | Context |
|---|------------|------|----------|---------|
| 1 | Lady Serana Valdren | 121 | [[Lady Serana Valdren]] | "...wife, manages court..." |
| 2 | High Confessor Maren | 136 | [[High Confessor Maren]] | "...supporters, seeking..." |
| 3 | Sister Elspeth | 204 | [[Sister Elspeth]] | "...Abbess of the..." |
| 4 | Lord Varic | 117 | [[Lord Varic Valdren|Lord Varic]] | "...ensuring Varic holds..." |

---

PARTIAL MATCHES (display text needed):

| # | Text Found | Line | Suggested Link |
|---|------------|------|----------------|
| 1 | "the Owl" | 123 | [[The Owl|the Owl]] |
| 2 | "young Edric" | 117, 177 | [[Edric Valdren|young Edric]] |

---

AMBIGUOUS (multiple possible matches):

| # | Text Found | Line | Could Match |
|---|------------|------|-------------|
| 1 | "Edric" | 177 | [[Edric Valdren]] or [[Prince Edric the Liberator]] |

---

SUMMARY:
- Exact matches: X
- Partial matches: X
- Ambiguous: X
- Total changes: X

Options:
1. Apply all exact matches
2. Apply exact + partial matches
3. Review each individually
4. Dry run complete (--dry-run flag)

If --dry-run flag is set, show report and stop.

Step 7: Apply Changes

For each approved match:

  1. Build replacement text:

- Exact match: EntityName → [[EntityName]] - Partial match: PartialName → [[FullEntityName|PartialName]]

  1. Apply using Edit tool:

- Process from bottom of file to top (so line numbers don't shift) - Use precise string replacement with surrounding context for uniqueness - Verify each replacement succeeded

  1. Handle duplicates:

- If the same entity is mentioned multiple times, link ALL occurrences - Exception: Don't link the same name twice in the same sentence

Step 8: Validation

After applying changes:

  1. Re-read the file
  2. Verify all intended wikilinks are present
  3. Check no content was accidentally corrupted
  4. Verify wikilinks point to existing files

Step 9: Summary Report

=== LINKIFY COMPLETE: [Entity Name] ===

Changes Applied: X

Links Added:
- [[Lady Serana Valdren]] (line 121)
- [[High Confessor Maren]] (line 136)
- [[Sister Elspeth]] (lines 204, 252)
- [[Lord Varic Valdren|Lord Varic]] (line 117)

Skipped (ambiguous):
- "Edric" (line 177) - multiple matches possible

---

File Updated: [path/to/entity.md]

Suggested Next Steps:
- Review ambiguous matches manually
- Run /audit-world to check bidirectional links
- Use /link-entities to add reciprocal connections

Matching Rules

Word Boundary Matching

Only match complete words/phrases:

TextEntityMatch?Reason
"spoke to Lord Varic Valdren about"Lord Varic ValdrenYesComplete phrase
"the Valdren family"House ValdrenNo"Valdren" alone, different entity
"visited Aldersgate"AldersgateYesComplete word
"Aldersgates"AldersgateNoDifferent word
"[[Lord Varic Valdren]]"Lord Varic ValdrenNoAlready linked

Partial Name Handling

When text contains a shorter form of an entity name:

TextEntityLink Format
"Lord Varic said"Lord Varic Valdren`[[Lord Varic Valdren
"the Owl knows"The Owl`[[The Owl
"young Edric"Edric Valdren`[[Edric Valdren

Case Handling

By default (case-insensitive):

  • "lord varic valdren" matches "Lord Varic Valdren"
  • Original case is preserved in display text

With --case-sensitive:

  • Only exact case matches

Exclusion Patterns

Never link:

  • Text inside existing [[wikilinks]]
  • Text in YAML frontmatter
  • Text in code blocks
  • Text in Image Prompts section
  • The entity's own name (no self-linking)
  • Text that's part of a markdown header (# Name)
  • Text in the Connections section (manage links there separately)

Bulk Processing Mode

When --world flag is provided, the skill processes multiple entities:

Bulk Mode Workflow

  1. Scan world directory:

- List all .md files in Worlds/[World Name]/ - If --category specified, only scan that folder (e.g., Characters/) - Exclude World Overview.md by default (use --include-overview to include)

  1. Build combined entity index:

- Same process as single-entity mode - Index is built once and reused for all files

  1. Process each entity:

- For each file, run Steps 4-7 (Find Matches → Apply Changes) - Track cumulative statistics

  1. Batch report: === BULK LINKIFY COMPLETE: [World Name] === Files Processed: X Files Modified: Y Files Unchanged: Z Total Links Added: N By Category: - Characters: X files, Y links - Settlements: X files, Y links - Organizations: X files, Y links... Skipped (ambiguous): N references across M files Performance: X.X seconds total

Bulk Mode Options

# Process entire world
/linkify --world Eldermyr

# Process only one category
/linkify --world Eldermyr --category Settlements

# Dry run to preview bulk changes
/linkify --world Eldermyr --dry-run

# Process multiple categories
/linkify --world Eldermyr --category Characters --category Organizations

Performance Notes

For large worlds (100+ entities), bulk mode:

  • Builds the entity index once (faster than per-file)
  • Processes files in parallel where possible
  • Reports progress every 10 files
  • For very large worlds (300+), consider using scripts/linkify_world.py instead

Examples

# Basic linkify
/linkify "Aldersgate"

# Preview what would be linked
/linkify "Aldersgate" --dry-run

# Linkify checking all worlds for entities
/linkify "Aldersgate" --all-worlds

# Linkify by path
/linkify Worlds/Eldermyr/Settlements/Aldersgate.md

# Case-sensitive matching
/linkify "Lord Varic Valdren" --case-sensitive

Integration with Other Skills

After /create-entity

Run /linkify on newly created entities to link any mentioned existing entities.

Before /audit-world

Run /linkify on key entities to add missing links before the audit.

With /link-entities

  • /linkify adds wikilinks within entity content
  • /link-entities manages the Connections section relationships
  • Use together for comprehensive linking

With /populate-entity

  • /linkify only links to EXISTING entities
  • /populate-entity --links-only does the same thing
  • /populate-entity (full) also CREATES missing entities
  • Use /linkify when you just want to add links without creating anything

Error Handling

File not found:

Error: Could not find entity "[name]"
Similar entities in Worlds/:
- [similar1]
- [similar2]
Please specify the full path or exact name.

No matches found:

=== LINKIFY: [Entity Name] ===

No unlinked entity references found.

This file either:
- Already has all entities properly linked
- Contains no references to other known entities
- References entities that don't exist yet (use /populate-entity to create them)

Write permission error:

Error: Could not write to [path]
The dry-run showed X matches. Please check file permissions.

平台分布

mcpjam

27.4%
按下载量换算18

Claude Code

25.48%
按下载量换算17

windsurf

16.7%
按下载量换算11

zencoder

13.81%
按下载量换算9

crush

7.59%
按下载量换算5

cline

3.48%
按下载量换算2

安全审计

暂无安全审计结果可展示。

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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