Cyber Library
Search and read Cyber Library, a free structured cybersecurity knowledge base. Read-only, no auth.
我該用這個嗎
品質與安全性
發現項目（2）
- HIGHTool poisoning patterns detected
- MEDIUMTool description contains suspicious base64-like encoded string在 get_page 中
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高，可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案：
{
"mcpServers": {
"library": {
"url": "https://www.cyberlibrary.com/api/mcp"
}
}
}遠端端點
https://www.cyberlibrary.com/api/mcpstreamable-http它能做什麼
工具清單
工具（6）
🟢search_library(query, locale, limit)
Full-text search over the published pages of cyberlibrary.com (titles, descriptions, paths). Returns the best matches with their path and URL; pass a path to get_page to read it. Docs: https://www.cyberlibrary.com/en/mcp
輸入結構描述
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 2,
"description": "Search terms, e.g. \"kerberoasting\" or \"container escape\""
},
"locale": {
"description": "Locale code, default \"en\"",
"type": "string"
},
"limit": {
"description": "Max results, default 10",
"type": "integer",
"minimum": 1,
"maximum": 50
}
},
"required": [
"query"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_page(path, locale, max_chars)
Returns a page of the Cyber Library as markdown with its metadata (title, URL, breadcrumbs, last update, contributors) and, for folders, the list of child pages. Paths look like "offensive/software/applications/server/containers/container-escape"; full URLs are accepted. Docs: https://www.cyberlibrary.com/en/mcp
輸入結構描述
{
"type": "object",
"properties": {
"path": {
"type": "string",
"minLength": 1,
"description": "Page path or URL"
},
"locale": {
"description": "Locale code, default \"en\"",
"type": "string"
},
"max_chars": {
"description": "Truncate the body after this many characters (default: no limit)",
"type": "integer",
"minimum": 500,
"maximum": 200000
}
},
"required": [
"path"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢list_topics(path, locale)
Lists the direct children of a folder of the library (or the six top-level sections when no path is given): title, path, number of sub-pages and description. Docs: https://www.cyberlibrary.com/en/mcp
輸入結構描述
{
"type": "object",
"properties": {
"path": {
"description": "Folder path, e.g. \"offensive/software\"; omit for the top level",
"type": "string"
},
"locale": {
"description": "Locale code, default \"en\"",
"type": "string"
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_outline(path, depth, locale)
Indented outline of the library tree below a path (default: whole library) down to a given depth. Useful to understand how a domain is organized before reading pages. Capped at 400 entries. Docs: https://www.cyberlibrary.com/en/mcp
輸入結構描述
{
"type": "object",
"properties": {
"path": {
"description": "Branch path; omit for the whole library",
"type": "string"
},
"depth": {
"description": "Levels to include, default 2",
"type": "integer",
"minimum": 1,
"maximum": 6
},
"locale": {
"description": "Locale code, default \"en\"",
"type": "string"
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢search_planned_topics(query, limit, status)
Searches the structure database snapshot of the Cyber taxonomy: every planned topic (8k+), including drafts not yet published on the site, with its status (draft, review, ready, generated, published) and position. Use get_planned_subtree with an id to explore around a match. Docs: https://www.cyberlibrary.com/en/mcp
輸入結構描述
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 2,
"description": "Title search terms"
},
"limit": {
"description": "Max results, default 15",
"type": "integer",
"minimum": 1,
"maximum": 50
},
"status": {
"description": "Only topics with this status",
"type": "string",
"enum": [
"draft",
"review",
"ready",
"generated",
"published"
]
}
},
"required": [
"query"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_planned_subtree(topic_id, depth)
Outline of the planned taxonomy below a topic id (from search_planned_topics; omit for the Cyber root) with each topic's kind and status. Capped at 400 entries. Docs: https://www.cyberlibrary.com/en/mcp
輸入結構描述
{
"type": "object",
"properties": {
"topic_id": {
"description": "Topic id; omit for the Cyber root",
"type": "integer",
"minimum": -9007199254740991,
"maximum": 9007199254740991
},
"depth": {
"description": "Levels to include, default 2",
"type": "integer",
"minimum": 1,
"maximum": 8
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}社群
證據