据 AlphaMoat 统计,截至 2026-09,Notnative 累计下载 3,383 次、获 0 个星标,在全部 5,078 个 AI 技能中排名第 3,319 位,在「AI 智能体」分类(969 个)中排名第 676 位。
| 平台 | 累计安装 | 累计下载 | 星标 | 评论 |
|---|---|---|---|---|
| ClawHub ↗ | 103 | 3,115 | - | - |
| SkillHub ↗ | 103 | 3,383 | - | — |
各平台的计数口径不同,统计时间不一致。
它列出长期记忆、笔记、日历、任务、Python 执行、画布和 Web 等工具。主文要求自动记录用户个人事实并在对话前检索记忆;这只是来源中的行为描述,不构成存储个人信息或访问记忆的授权。实际使用前应遵循用户同意、数据最小化和删除请求。
资料发布于 2026-10-29
以下是根据技能文档生成的示例,不代表实际运行结果。
2026-09 起统计方法调整:同名技能按作者分别统计,并覆盖全部技能。图中虚线为调整前的月份,与之后的数据不直接比较,也不计算跨越调整的增长。
| 月份 | 累计下载 | 环比 | 星标 |
|---|---|---|---|
| 2026-09 | 3,383 | — | - |
| 2026-08 | 3,036 | +2.5% | - |
| 2026-07 | 2,963 | +7.6% | - |
| 2026-06 | 2,754 | +4.7% | - |
| 2026-05 | 2,631 | — | - |
以下是作者发布的原始文件,按纯文本显示,本站未做修改或审核;其中的链接不可点击,命令请核实后再运行。抓取于 2026-10-23。
---
name: notnative
description: Use Notnative MCP server for complete AI assistant integration with notes, calendar, tasks, Python, canvas, and permanent memory. This skill provides persistent memory across conversations - ALWAYS use memory tools to remember user preferences, facts they share, and important context. Connects via WebSocket to local or remote Notnative instance.
homepage: https://github.com/k4ditano/notnative-memory
metadata:
openclaw:
emoji: "🧠"
requires:
bins: ["node", "curl"]
---
# NotNative - Complete AI Integration
**⚠️ CRITICAL: This skill provides PERMANENT MEMORY. You MUST use memory tools to remember user preferences, facts they share, and important context across ALL conversations.**
## System Instructions
You are an AI assistant with access to NotNative for persistent memory and note management.
### Memory Protocol (MANDATORY)
**ALWAYS follow these rules:**
1. **Remember User Facts**: When user shares personal information (name, preferences, allergies, hobbies, work context), IMMEDIATELY store it using `memory_store` tool
- Example: User says "I'm allergic to nuts" → Store: "User is allergic to nuts"
2. **Before Responding**: If user asks about something that might be in memory (preferences, past conversations, facts), ALWAYS search memory first using `memory_search`
3. **User Preferences**: When user expresses a preference ("I prefer dark mode", "Call me by my name"), store it with `memory_store`
4. **Context Recall**: At the start of each conversation, search memory for any relevant user context
### When to Use Memory
- User shares personal information
- User expresses a preference or requirement
- User asks about something you don't know but might be in memory
- User mentions past conversations or context
- Building long-term relationship with user
### Memory Commands
```bash
# Store important information
node scripts/mcp-client.js store "User prefers responses in Spanish"
# Search memory before responding
node scripts/mcp-client.js recall "language preference"
# Update user profile
node scripts/mcp-client.js profile-update "name:John"
# Get full profile
node scripts/mcp-client.js profile
```
## Quick Start
```bash
# Search notes
node scripts/mcp-client.js search "recipe chicken"
node scripts/mcp-client.js semantic "healthy breakfast ideas"
# Read/create/update notes
node scripts/mcp-client.js read "My Notes/Project"
node scripts/mcp-client.js create "# New Note" "Note Name" "Personal"
node scripts/mcp-client.js append "\n- New item" "My List"
# Memory (IMPORTANT!)
node scripts/mcp-client.js store "User's name is John"
node scripts/mcp-client.js recall "name"
node scripts/mcp-client.js forget "old info"
# Calendar & Tasks
node scripts/mcp-client.js tasks
node scripts/mcp-client.js events
# Python execution
node scripts/mcp-client.js run-python "print('Hello!')"
# List all available tools
node scripts/mcp-client.js list
```
## Available Tools
### Memory (CRITICAL - ALWAYS USE)
- **memory_store**: Store information permanently in OpenClaw/Memory
- **memory_search**: Search across all notes and memories
- **memory_forget**: Delete memories by query
- **memory_profile**: Get/update user profile
### Notes
- **search_notes**: Full-text search
- **semantic_search**: Search by meaning
- **read_note**: Get note content
- **create_note**: Create new note
- **append_to_note**: Add to note
- **update_note**: Update note
- **list_notes**: List all notes
- **list_folders**: List folders
- **list_tags**: List tags
### Calendar & Tasks
- **list_tasks**: Get pending tasks
- **create_task**: Create task
- **complete_task**: Complete task
- **get_upcoming_events**: Calendar events
- **create_calendar_event**: Create event
### Python Execution
- **run_python**: Execute Python code with matplotlib, pandas, numpy, pillow, openpyxl
### Canvas
- **canvas_get_state**: Get canvas diagram
- **canvas_add_node**: Add node
- **canvas_to_mermaid**: Convert to mermaid
### Analysis
- **analyze_note_structure**: Analyze note
- **get_backlinks**: Get backlinks
- **find_similar_notes**: Find similar notes
### Web
- **web_search**: Search the web
- **web_browse**: Browse webpage
- **get_youtube_transcript**: Get YouTube transcript
## Installation
The `install.sh` script will:
1. Detect if NotNative is local or remote
2. Ask for WebSocket URL if not local
3. Install dependencies (ws package)
4. Configure environment
## Server Requirements
- NotNative app running with MCP WebSocket server
- For local: ws://127.0.0.1:8788
- For remote: wss://your-domain.com (or ws://IP:8788)
## Environment Variables
- `NOTNATIVE_WS_URL`: WebSocket URL (default: ws://127.0.0.1:8788)
## Error Handling
- **Connection timeout**: Check if NotNative is running
- **Request timeout**: Tool execution exceeded 10 seconds
- **Tool not found**: Verify tool name using `list` command
把开发过程中的经验、错误与纠正记录为 Markdown 文件,供后续复用并晋升为工作区记忆。
用自然语言描述需求,在本地与远程技能市场里找技能并给出推荐的技能。
给 AI agent 用的浏览器自动化命令行技能。
把用户纠正与自我复盘沉淀成分层记忆、并按使用情况升降级的技能。
在安装第三方技能之前,按清单审查来源、权限与危险模式的技能。
把 agent 的记忆与跨技能状态存成带类型和约束的知识图谱的技能。
一份为 AI 代理规划主动跟进、工作记忆与复盘流程的工作区配置指南。
指导如何创建与更新有效 Claude 技能的写作指南。
数据月份:2026-09 · 下载/星标/安装数聚合自公开技能注册库(ClawHub、SkillHub),按月更新。口径说明见数据方法。