功能描述

Search and fetch articles from Grokipedia.com — xAI's AI-generated encyclopedia (like Wikipedia but written by Grok). Use when asked about topics that might have a Grokipedia article, or when the user explicitly mentions Grokipedia.

使用说明 (SKILL.md)

Grokipedia Parser

Name: Grokipedia
Author: kirillleventcov

Search and fetch articles from Grokipedia.com — xAI's AI-generated encyclopedia.

Source: github.com/kirillleventcov/grokipedia-parser

Requirements

Node.js (v18+) / Bun — used to run the search and fetch scripts
Dependencies — jsdom and @mozilla/readability (installed via bun install)

Install

cd ~/.openclaw/workspace/skills/Grokipedia
bun install --production

Note: Installation creates a node_modules/ directory in the skill folder. The scripts themselves only output to stdout at runtime.

Scripts

Search Articles

node ~/.openclaw/workspace/skills/Grokipedia/scripts/search.mjs "query" [--limit N]

Parameters:

query - Search term (required)
--limit N - Max results (1-50, default: 10)

Output: JSON array with slug, title, snippet, relevanceScore

Example:

node ~/.openclaw/workspace/skills/Grokipedia/scripts/search.mjs "artificial intelligence" --limit 5

Fetch Article

node ~/.openclaw/workspace/skills/Grokipedia/scripts/fetch.mjs "Article_Slug"

Parameters:

slug - Article slug (required, case-sensitive, use underscores)

Output: Clean markdown article content

Example:

node ~/.openclaw/workspace/skills/Grokipedia/scripts/fetch.mjs "Helsinki"
node ~/.openclaw/workspace/skills/Grokipedia/scripts/fetch.mjs "Artificial_intelligence"

What This Skill Does

Network access: Fetches from grokipedia.com only (search API + article pages)
No credentials: Public read-only access, no API keys or tokens needed
No runtime file writes: Only outputs to stdout (JSON for search, markdown for articles). Install step creates node_modules/ in the skill directory.
No persistence: No background processes, no cron, no elevated privileges
Dependencies: jsdom (DOM parsing) and @mozilla/readability (article extraction)

Notes

Article slugs are case-sensitive (e.g., Helsinki not helsinki)
Slugs use underscores for spaces (e.g., Artificial_intelligence)
Search returns up to 50 results max
Articles contain internal links in format [text](/page/Slug)
Content is AI-generated by Grok

安全使用建议

This skill appears to do exactly what it says: perform read-only search/fetch operations against grokipedia.com and return cleaned markdown/JSON. Before installing, note that 'bun install' will fetch npm packages and create a node_modules directory inside the skill folder (inspect these dependencies if you require strict supply-chain controls). The scripts make outbound HTTPS requests to grokipedia.com — if you need to restrict network access, run the skill in a network-restricted or sandboxed environment. If you have policies about trusting AI-generated content, remember Grokipedia content is AI-generated and should be validated independently.

功能分析

Type: OpenClaw Skill Name: grokipedia Version: 1.2.0 The skill bundle is designed to search and fetch articles from grokipedia.com, aligning with its stated purpose. Code analysis of `scripts/fetch.mjs` and `scripts/search.mjs` shows proper input validation (e.g., `slug` and `query` sanitization, length limits) and safe URL construction using `encodeURIComponent` or `URL.searchParams.set`. All network requests are explicitly directed to `grokipedia.com`. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection attempts in `SKILL.md` or `README.md`. Dependencies are standard for HTML parsing. The `--raw` HTML output option in `fetch.mjs` is a feature, and while it could pose an XSS risk if rendered unsafely by the consuming agent, this is a vulnerability in the consumer, not the skill itself.

能力评估

✓ Purpose & Capability

The name/description say 'search and fetch from grokipedia.com' and the code (search.mjs, fetch.mjs) only calls grokipedia API/page endpoints. Declared dependencies (jsdom, @mozilla/readability) and Node/Bun runtime are appropriate for HTML parsing and extraction.

✓ Instruction Scope

SKILL.md and the scripts limit runtime actions to performing HTTPS requests to grokipedia.com, parsing results, and writing output to stdout. The scripts validate inputs (slug and query) and do not read system files or environment variables beyond process.argv.

ℹ Install Mechanism

There is no custom installer; SKILL.md recommends running 'bun install --production' which will create node_modules in the skill folder and fetch packages from the registry. This is expected but means network access to package registries during install and code will be written to disk (node_modules). No arbitrary binary downloads or obscure URLs are used.

✓ Credentials

The skill declares no required environment variables, credentials, or config paths. The code likewise does not access secrets or unrelated environment variables.

✓ Persistence & Privilege

The skill does not request 'always' presence, does not spawn background processes, and does not modify other skills. It can be invoked autonomously (default platform behavior), which is normal and not excessive here.

版本历史

v1.2.0

- Added support for Bun as an alternative to Node.js for installing and running dependencies. - Updated install instructions to use bun and clarified that installation creates a node_modules/ directory in the skill folder. - Adjusted documentation paths to reflect the new skill directory structure under ~/.openclaw/workspace/skills/Grokipedia. - Clarified that only the install step writes files, and scripts themselves do not write files at runtime. - Minor cleanup and clarification in requirements and usage notes.

v1.1.1

- Added a `name` and `description` section in SKILL.md with usage guidance. - No code changes in this release; documentation improved for clarity and discoverability.

v1.1.0

- Added openclaw.summary to package.json (fixes 'No summary provided') - Added openclaw.requiredBinaries: ['node'] (declares runtime dependency) - Added openclaw.install step for npm dependencies - Added homepage/repository/license to package.json - SKILL.md now documents requirements, install steps, network access scope, and explicit 'no file writes / no credentials / no persistence' guarantees

v1.0.0

Initial release of grokipedia skill. - Enables searching Grokipedia.com with customizable result limits. - Allows fetching full articles by slug in clean markdown format. - Provides scripts for easy access: search and fetch. - Returns search results with relevance score and snippets. - Handles Grokipedia-specific slugs (case-sensitive, underscores for spaces).

元数据

Slug grokipedia

版本 1.2.0

许可证 —

累计安装 3

当前安装数 3

历史版本数 4

常见问题

Grokipedia 是什么？

Search and fetch articles from Grokipedia.com — xAI's AI-generated encyclopedia (like Wikipedia but written by Grok). Use when asked about topics that might have a Grokipedia article, or when the user explicitly mentions Grokipedia. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 2110 次。

如何安装 Grokipedia？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install grokipedia」即可一键安装，无需额外配置。

Grokipedia 是免费的吗？

是的，Grokipedia 完全免费（开源免费），可自由下载、安装和使用。

Grokipedia 支持哪些平台？

Grokipedia 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 Grokipedia？

由 kirillleventcov（@kirillleventcov）开发并维护，当前版本 v1.2.0。

Grokipedia