Token导航 LogoToken导航

Agent Skill · 研究检索

notion-master(Notion master 搜索)

用于处理 Notion 页面、数据库、工作区内容和结构化记录。它适合让 Agent 查询知识库、整理页面内容、创建记录或把外部信息同步到 Notion。使用时需要确认集成是否已被授权到目标页面或数据库,并区分读取、追加和覆盖更新;涉及批量写入或修改数据库属性时,应先核对字段名称、属性类型和目标页面。

notion-master 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,权限为接触敏感数据;许可证为 MIT,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 58,786。

总安装 404周安装 17GitHub Stars 2下载 141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:notion-master(Notion master 搜索)
来源仓库:https://github.com/abdullahbeam/nexus-design-abdullah
仓库路径:skills/notion-master
安装命令:
npx skills add https://github.com/abdullahbeam/nexus-design-abdullah --skill notion-master
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/abdullahbeam/nexus-design-abdullah --skill notion-master

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。

用于处理 Notion 页面、数据库、工作区内容和结构化记录。适合让 Agent 查询知识库、整理页面内容、创建记录或把外部信息同步到 Notion。使用时需要确认集成是否已被授权到目标页面或数据库,并区分读取、追加和覆盖更新。

安装命令:npx skills add https://github.com/abdullahbeam/nexus-design-abdullah --skill notion-master。涉及批量写入或修改数据库属性时,应先核对字段名称、属性类型和目标页面。

来源仓库内路径:skills/notion-master

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/abdullahbeam/nexus-design-abdullah --skill not…;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

    它需要接触敏感数据,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

notion-master 怎么安装?

在终端执行 npx skills add https://github.com/abdullahbeam/nexus-design-abdullah --skill notion-master。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

notion-master 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。

notion-master 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

notion-master 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。

notion-master 用的人多吗?

公开数据是累计安装 404 次,近一周安装 17 次,下载 141 次,GitHub 2 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。

notion-master 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 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

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 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
notion-master

Notion master 搜索 · 本页

4042接触敏感数据MIT未通过

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

Notion Master

This is NOT a user-facing skill. It's a shared resource library referenced by Notion integration skills.

Purpose

Provides shared resources to eliminate duplication across:

  • notion-connect - Meta-skill for complete Notion workspace integration (NEW)
  • query-notion-db - Browse and search Notion skills database
  • import-skill-to-nexus - Download skills from Notion to local Nexus
  • export-skill-to-notion - Push skills to Notion for team sharing

Instead of loading this skill, users directly invoke the specific skill they need above.


Architecture: DRY Principle

Problem solved: The 3 Notion skills had 950 lines of duplicated content (setup wizard 3x, API docs 3x, error tables 3x).

Solution: Extract shared content into notion-master/references/ and notion-master/scripts/, then reference from each skill.

Result: 60% context reduction (950 → 370 lines in SKILL.md files)


Shared Resources

All 3 Notion skills reference these resources (progressive disclosure).

references/

setup-guide.md - Complete setup wizard

  • Getting Notion API key
  • Finding database ID
  • Configuring.env and user-config.yaml
  • Getting user's Notion ID

api-reference.md - Notion API patterns

  • Database query API
  • File upload API (3-step process)
  • File download API (3-step process)
  • Common curl examples

error-handling.md - Troubleshooting

  • Common errors and solutions
  • API error codes
  • Configuration issues
  • Network and timeout handling

database-schema.md - Beam Nexus Skills database

  • Property types and descriptions
  • Available Teams and Integrations
  • Field mapping table
  • Example entries

filter-syntax.md - Query filter syntax

  • Operators by property type
  • Common filter patterns
  • Sorting options

property-types.md - Property type reference

  • All 20+ property types
  • Read and write formats
  • Validation rules

block-types.md - Block type reference

  • All 50+ block types
  • Block schemas and examples
  • Rich text formatting

scripts/

Configuration & Setup

check_notion_config.py - Pre-flight validation

python check_notion_config.py [--json]
ArgumentRequiredDefaultDescription
--jsonNoFalseOutput structured JSON for AI consumption

Exit codes: 0=configured, 1=partial, 2=not configured

When to Use: Run this FIRST before any Notion operation. Use to validate API key is configured, check if Skills DB ID is set, diagnose connection issues, or determine if setup wizard is needed.


setup_notion.py - Interactive setup wizard

python setup_notion.py

No arguments - runs interactively. Guides through API key setup, tests connection, saves to .env, gets user's Notion ID, auto-runs database discovery.

When to Use: Use when Notion integration needs initial setup, when check_notion_config.py returns exit code 2, or when user needs to reconfigure credentials or change integration settings.


discover_databases.py - Database discovery (POST /search)

python discover_databases.py [--refresh] [--json] [--detailed]
ArgumentRequiredDefaultDescription
--refreshNoFalseForce refresh even if context file exists
--jsonNoFalseOutput JSON only (no file save)
--detailedNoFalseFetch full schema for each database (slower)

Saves to: 01-memory/integrations/notion-databases.yaml

When to Use: Use when user asks "what databases do I have", "list my notion databases", after connecting a new integration, or when database name resolution fails (--refresh to update cache).


Database Querying

search_skill_database.py - Unified database querying (POST /databases/{id}/query)

# General mode
python search_skill_database.py --db "Database Name" [--filter EXPR] [--sort FIELD] [--sort-dir asc|desc] [--limit N] [--json]

# Skills DB preset mode
python search_skill_database.py --skills [--team TEAM] [--integration INT] [--name TEXT] [--owner ID] [--json]
ArgumentRequiredDefaultDescription
--dbNo*-Database name or ID (fuzzy match)
--skillsNo*FalseQuery Beam Nexus Skills database (preset mode)
--teamNo-[Skills mode] Filter by team (General, Solutions, etc.)
--integrationNo-[Skills mode] Filter by integration (Beam AI, Linear, etc.)
--nameNo-[Skills mode] Filter by skill name (partial match)
--ownerNo-[Skills mode] Filter by owner user ID
--filterNo-Filter expression (see syntax below)
--sortNo-Property name to sort by
--sort-dirNodescSort direction: asc or desc
--limitNo-Limit number of results
--jsonNoFalseOutput raw JSON

*Either --db or --skills is required

Filter Syntax:

--filter "Status = Active"           # Equals
--filter "Status != Done"            # Not equals
--filter "Name contains project"     # Contains
--filter "Priority > 5"              # Greater than
--filter "Due Date < 2025-01-01"     # Less than
--filter "Tags is_empty"             # Is empty
--filter "Tags is_not_empty"         # Is not empty

When to Use: Use when user wants to query any Notion database, search for records, filter by properties, find skills in the Skills DB (--skills mode), or list database contents with sorting/pagination.


Page Operations

create_page.py - Create page in database (POST /pages)

python create_page.py --db "Database" [--properties JSON] [--interactive] [--json]
ArgumentRequiredDefaultDescription
--dbYes-Database name or ID (fuzzy match)
--propertiesNo-JSON object with property values
--interactiveNoFalsePrompt for each property
--jsonNoFalseOutput raw JSON

When to Use: Use when user wants to add a new entry to a database, create a new record, or add an item to a Notion table. Use --interactive for guided property entry.


manage_page.py - Page management (GET/PATCH /pages/{id})

python manage_page.py get --page PAGE_ID [--json]
python manage_page.py update --page PAGE_ID --properties JSON [--json]
python manage_page.py delete --page PAGE_ID [--confirm] [--json]

Subcommand: get

ArgumentRequiredDefaultDescription
--pageYes-Page ID
--jsonNoFalseOutput raw JSON

Subcommand: update

ArgumentRequiredDefaultDescription
--pageYes-Page ID
--propertiesYes-JSON object with property values
--jsonNoFalseOutput raw JSON

Subcommand: delete

ArgumentRequiredDefaultDescription
--pageYes-Page ID
--confirmNoFalseSkip confirmation prompt
--jsonNoFalseOutput raw JSON

When to Use: Use get to retrieve page details/properties, update to modify page properties (status, tags, dates), delete to archive/remove pages. Requires page ID from query results.


Database Management

manage_database.py - Database operations (POST/PATCH /databases)

python manage_database.py create --parent PAGE_ID --title "Title" [--properties JSON] [--inline] [--json]
python manage_database.py update --db DB_ID [--title TEXT] [--add-property JSON] [--update-property JSON] [--json]

Subcommand: create

ArgumentRequiredDefaultDescription
--parentYes-Parent page ID
--titleYes-Database title
--propertiesNo[]JSON array of property definitions
--inlineNoFalseCreate inline database
--jsonNoFalseOutput raw JSON

Subcommand: update

ArgumentRequiredDefaultDescription
--dbYes-Database ID
--titleNo-New title
--add-propertyNo-JSON property definition to add
--update-propertyNo-JSON property update
--jsonNoFalseOutput raw JSON

When to Use: Use create to make a new database inside a page, update to add/modify database properties (columns), change database title, or add new select options.


Block Operations

manage_blocks.py - Block operations (GET/PATCH/DELETE /blocks/{id})

python manage_blocks.py get --block BLOCK_ID [--json]
python manage_blocks.py children --block BLOCK_ID [--limit N] [--all] [--json]
python manage_blocks.py append --block BLOCK_ID [--content JSON] [--type TYPE] [--text TEXT] [--language LANG] [--checked] [--json]
python manage_blocks.py update --block BLOCK_ID --content JSON [--json]
python manage_blocks.py delete --block BLOCK_ID [--confirm] [--json]

Subcommand: get

ArgumentRequiredDefaultDescription
--blockYes-Block ID
--jsonNoFalseOutput raw JSON

Subcommand: children

ArgumentRequiredDefaultDescription
--block / --pageYes-Block or page ID
--limitNo50Max blocks to return
--allNoFalseFetch all pages
--jsonNoFalseOutput raw JSON

Subcommand: append

ArgumentRequiredDefaultDescription
--block / --pageYes-Parent block or page ID
--contentNo-JSON block content
--typeNo-Block type (paragraph, heading_1, etc.)
--textNo-Text content
--languageNo-Code block language
--checkedNoFalseFor to_do blocks
--jsonNoFalseOutput raw JSON

Subcommand: update

ArgumentRequiredDefaultDescription
--blockYes-Block ID
--contentYes-JSON block content
--jsonNoFalseOutput raw JSON

Subcommand: delete

ArgumentRequiredDefaultDescription
--blockYes-Block ID
--confirmNoFalseSkip confirmation
--jsonNoFalseOutput raw JSON

When to Use: Use get to retrieve block info, children to read page content, append to add text/headings/code blocks to pages, update to modify existing blocks, delete to remove content.


User Management

manage_users.py - User operations (GET /users)

python manage_users.py list [--limit N] [--all] [--save] [--json]
python manage_users.py get --user USER_ID [--json]
python manage_users.py me [--json]

Subcommand: list

ArgumentRequiredDefaultDescription
--limitNo100Max users to return
--allNoFalseFetch all pages
--saveNoFalseSave users to context file
--jsonNoFalseOutput raw JSON

Subcommand: get

ArgumentRequiredDefaultDescription
--userYes-User ID
--jsonNoFalseOutput raw JSON

Subcommand: me

ArgumentRequiredDefaultDescription
--jsonNoFalseOutput raw JSON

When to Use: Use list to see all workspace members, get to lookup a specific user, me to get current authenticated user/bot identity (useful for setting notion_user_id during setup).


Comment Operations

manage_comments.py - Comment operations (GET/POST /comments)

python manage_comments.py list --block BLOCK_ID [--limit N] [--all] [--json]
python manage_comments.py create --page PAGE_ID --text "Comment text" [--json]
python manage_comments.py create --discussion THREAD_ID --text "Reply text" [--json]
python manage_comments.py get --comment COMMENT_ID [--json]

Subcommand: list

ArgumentRequiredDefaultDescription
--block / --pageYes-Block or page ID
--limitNo50Max comments
--allNoFalseFetch all pages
--jsonNoFalseOutput raw JSON

Subcommand: create

ArgumentRequiredDefaultDescription
--pageNo*-Page ID for new comment
--discussionNo*-Discussion thread ID for reply
--textYes-Comment text
--jsonNoFalseOutput raw JSON

*Either --page or --discussion is required

Subcommand: get

ArgumentRequiredDefaultDescription
--commentYes-Comment ID
--jsonNoFalseOutput raw JSON

Note: Requires "Insert comments" capability in integration settings.

When to Use: Use list to read comments on a page, create to add comments or reply to threads, get to retrieve specific comment details. Great for collaboration and feedback workflows.


Skill Import/Export

download_skill.py - Download skill from Notion

python download_skill.py PAGE_ID [PAGE_ID2 ...] [--output-dir PATH] [--no-backup] [--json]
ArgumentRequiredDefaultDescription
page_idsYes-One or more Notion page IDs (positional)
--output-dirNo03-skillsOutput directory for skill folder
--no-backupNoFalseDon't backup existing skills
--jsonNoFalseOutput result as JSON

When to Use: Use when user wants to import a skill from the Skills DB, download a skill bundle from Notion, or sync a skill from the team repository to local Nexus.


upload_skill.py - Upload skill to Notion

python upload_skill.py SKILL_PATH --team TEAM [--integration LIST] [--dry-run] [--as-new NAME] [--skip-validation] [--json]
ArgumentRequiredDefaultDescription
skill_pathYes-Path to skill folder (positional)
--teamYes-Team: General, Solutions, Engineering, Sales
--integrationNo-Comma-separated integrations (Beam AI, Linear, etc.)
--dry-runNoFalsePreview upload without pushing
--as-newNo-Upload as new skill with different name
--skip-validationNoFalseSkip SKILL.md validation
--jsonNoFalseOutput as JSON

When to Use: Use when user wants to share a local skill to the team, export a skill to Notion, or publish a skill to the Skills DB for others to use.


Utility Module

rate_limiter.py - Rate limit handling (module, not CLI)

from rate_limiter import with_retry, RateLimiter, make_request_with_retry

# Decorator usage
@with_retry(max_retries=3, base_delay=1.0)
def my_api_call():
    return requests.get(...)

# Direct usage
limiter = RateLimiter(max_retries=3, base_delay=1.0)
response = limiter.execute(lambda: requests.get(...))

Features: Exponential backoff, jitter, respects Retry-After header, auto-retry on 429/5xx

When to Use: This is a library module, not a CLI script. Import it in Python code when making direct API calls that need rate limit protection (3 req/s limit).


Intelligent Error Detection Flow

When a Notion skill fails due to missing configuration, the AI should:

Step 1: Run Config Check with JSON Output

python 00-system/skills/notion/notion-master/scripts/check_notion_config.py --json

Step 2: Parse the ai_action Field

The JSON output includes an ai_action field that tells the AI what to do:

ai_actionWhat to Do
proceed_with_operationConfig OK, continue with the original operation
proceed_with_warningPartial config (can query/import but not export)
prompt_for_api_keyAsk user: "I need your Notion API key. Get one at https://www.notion.so/my-integrations"
create_env_fileCreate .env file and ask user for API key
run_setup_wizardRun: python 00-system/skills/notion/notion-master/scripts/setup_notion.py

Step 3: Help User Fix Issues

If ai_action is prompt_for_api_key:

  1. Tell user: "Notion integration needs setup. I need your API key."
  2. Show them: "Get one at: https://www.notion.so/my-integrations"
  3. Ask: "Paste your Notion API key here:"
  4. Once they provide it, write directly to .env: # Edit.env file to add: NOTION_API_KEY=secret_their_key_here NOTION_SKILLS_DB_ID=2bc2cadf-bbbc-80be-af8a-d45dfc8dfa2e
  5. Re-run config check to verify

JSON Output Structure

{
  "status": "not_configured",
  "exit_code": 2,
  "ai_action": "prompt_for_api_key",
  "missing": [
    {"item": "NOTION_API_KEY", "required": true, "location": ".env"}
  ],
  "fix_instructions": [...],
  "env_template": "NOTION_API_KEY=secret_YOUR_API_KEY_HERE\nNOTION_SKILLS_DB_ID=...",
  "setup_wizard": "python 00-system/skills/notion/notion-master/scripts/setup_notion.py"
}

How Skills Reference This

Each skill loads shared resources only when needed (progressive disclosure):

notion-connect uses:

  • All discovery and query scripts
  • All page, block, user, and comment management scripts
  • All references (filter-syntax, property-types, block-types)

query-notion-db uses:

  • check_notion_config.py (validate before query)
  • api-reference.md (query patterns)
  • error-handling.md (troubleshooting)

import-skill-to-nexus uses:

  • check_notion_config.py (validate before import)
  • api-reference.md (file download API)
  • error-handling.md (troubleshooting)

export-skill-to-notion uses:

  • check_notion_config.py (validate before export)
  • api-reference.md (file upload API)
  • database-schema.md (field mapping)
  • error-handling.md (troubleshooting)

Usage Example

User says: "query my Projects database"

What happens:

  1. AI loads notion-connect skill (NOT notion-master)
  2. notion-connect SKILL.md says: "Run check_notion_config.py first"
  3. AI executes: python notion-master/scripts/check_notion_config.py
  4. AI executes: python notion-master/scripts/search_skill_database.py --db "Projects"
  5. If errors occur, AI loads: notion-master/references/error-handling.md

notion-master is NEVER loaded directly - it's just a resource library.


Version: 2.6 Created: 2025-12-10 Updated: 2025-12-11 Status: Production Ready

Changelog:

  • v2.6: Added "When to Use" sections to all 13 scripts for AI routing guidance
  • v2.5: Added comprehensive script argument documentation with usage examples and argument tables
  • v2.4: Added Intelligent Error Detection Flow with --json support for AI-guided setup
  • v2.3: Renamed query_database.py to search_skill_database.py for clarity

平台分布

Claude Code

29.28%
按下载量换算41

Antigravity

21.26%
按下载量换算30

windsurf

15.98%
按下载量换算23

Codex

11.73%
按下载量换算17

OpenCode

6.84%
按下载量换算10

Gemini CLI

3.58%
按下载量换算5

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

未通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。

继续浏览同类 Skills