什么是 reMarkable 集成？

Openclaw Skills 的 reMarkable 集成在纸笔手写与数字 AI 智能体之间建立了无缝桥梁。通过利用 rmapi 云端 API，此技能允许您的智能体直接从平板电脑获取手写笔记、草图和图表。无论您是想将个人日记数字化还是增强粗略的绘图，该工具都能将类纸化体验转化为可操作的数字数据。将此集成到您的 Openclaw Skills 库中，可确保您的模拟头脑风暴会议能够立即为您的数字项目和记忆系统做出贡献。

下载入口:https://github.com/openclaw/skills/tree/main/skills/coolmanns/remarkable-tablet

安装与下载

1. ClawHub CLI

从源直接安装技能的最快方式。

2. 手动安装

将技能文件夹复制到以下位置之一

优先级：工作区 > 本地 > 内置

3. 提示词安装

将此提示词复制到 OpenClaw 即可自动安装。

reMarkable 集成 应用场景

• 通过获取手写感悟并将其附加到数字 Markdown 文件中，实现日记条目自动化。
• 使用 nano-banana-pro 等 AI 图像编辑工具增强粗略草图和图形。
• 从头脑风暴图表中提取结构和列表，以填充项目文档。
• 为博客文章和社交媒体内容美化手绘插图。

1. 在 reMarkable 平板上创建的内容将同步到 reMarkable 云端。
2. Openclaw Skills 智能体使用 rmapi 工具根据标签、星标或特定文件夹列出并查找文件。
3. 文档以 PDF 或 PNG 格式下载，并可选渲染注释以保证视觉保真度。
4. 智能体处理内容：使用视觉模型解释文本，而草图则被路由到图像增强流水线。
5. 最终输出要么存储在本地知识库中，要么作为优化后的资产返回给用户。

reMarkable 集成 配置指南

要开始使用 Openclaw Skills 的此组件，您必须安装 rmapi 工具并关联您的设备。

# 1. 下载 rmapi (ddvk 分支) 到 ~/bin/rmapi
# 2. 访问 https://my.remarkable.com/connect/desktop 获取连接码
# 3. 运行 rmapi 并输入 8 位字符代码进行身份验证

身份验证令牌存储在 ~/.rmapi 中，文件通常会同步到 ~/clawd/remarkable-sync/。

reMarkable 集成 数据架构与分类体系

该技能管理本地目录和云端元数据中的数据。信息组织如下：

name: remarkable
description: Fetch handwritten notes, sketches, and drawings from a reMarkable tablet via Cloud API (rmapi). Process content by refining artwork with AI image generation, extracting handwritten text to memory/journal, or using sketches as input for other workflows. Use when working with reMarkable tablet content, syncing handwritten notes, processing sketches, or integrating tablet drawings into projects.

reMarkable Tablet Integration (rmapi)

Fetch handwritten notes, sketches, and drawings from a reMarkable tablet via Cloud API. Process them — refine artwork with AI image generation, extract text to memory/journal, or use as input for other workflows.

Typical Use Cases

1. Journal entries — User writes thoughts on reMarkable → fetch → OCR/interpret → append to memory/YYYY-MM-DD.md or a dedicated journal file
2. Sketch refinement — User draws a rough graphic → fetch → enhance with nano-banana-pro (AI image editing) → return polished version
3. Brainstorming/notes — User jots down ideas, lists, diagrams → fetch → extract structure → add to project docs or memory
4. Illustrations — User creates hand-drawn art → fetch → optionally stylize → use in blog posts, social media, etc.

Processing Pipeline

reMarkable tablet → Cloud sync → rmapi fetch → PDF/PNG
                                                  ↓
                                    ┌─────────────┴─────────────┐
                                    │                           │
                              Text content?               Visual/sketch?
                                    │                           │
                              OCR / interpret            nano-banana-pro
                                    │                     (AI enhance)
                                    │                           │
                              Add to memory/            Return refined
                              journal/project            image to user

Setup

• Tool: rmapi (ddvk fork) v0.0.32
• Binary: ~/bin/rmapi
• Config: ~/.rmapi (device token after auth)
• Sync folder: ~/clawd/remarkable-sync/

Authentication (ONE-TIME)

1. User goes to https://my.remarkable.com/connect/desktop
2. Logs in, gets 8-character code
3. Run rmapi and enter the code
4. Token saved to ~/.rmapi — future runs are automatic

Commands

# List files/folders
rmapi ls
rmapi ls --json

# Navigate
rmapi cd "folder name"

# Find by tag / starred / regex
rmapi find --tag="share-with-gandalf" /
rmapi find --starred /
rmapi find / ".*sketch.*"

# Download (PDF)
rmapi get "filename"

# Download with annotations rendered (best for sketches)
rmapi geta "filename"

# Bulk download folder
rmapi mget -o ~/clawd/remarkable-sync/ "/Shared with Gandalf"

Sharing Workflows

Option A: Dedicated Folder

User creates "Shared with Gandalf" folder on reMarkable → moves items there → agent fetches with rmapi mget

Option B: Tag-Based

User tags documents with share-with-gandalf → agent discovers with rmapi find --tag

Option C: Starred Items

User stars items → agent fetches with rmapi find --starred

Fetch Script

# Fetch from shared folder
~/clawd/scripts/remarkable-fetch.sh

# Fetch starred items
~/clawd/scripts/remarkable-fetch.sh --starred

# Fetch by tag
~/clawd/scripts/remarkable-fetch.sh --tag="share-with-gandalf"

Notes

• Tablet must cloud-sync before files are available
• geta renders annotations into PDF (preferred for handwritten content)
• Use convert (ImageMagick) to go from PDF → PNG for image processing
• For text extraction, interpret the handwriting visually (vision model) rather than traditional OCR