@memoreru-sdk/cli

Memoreru CLI

日本語

Sync local Markdown, CSV, and JSON files with Memoreru — your content knowledge base.

Features

• Login — Browser-based login (CAPTCHA/2FA/OAuth supported)
• Keys — Create, list, and revoke API keys from the CLI
• Init — Generate project templates
• Pull — Download Memoreru content to local files
• Push — Upload local files to Memoreru
• Status — Show local changes since last sync
• Diff — Show file-level diffs before pushing
• Images — Auto-upload on push, download only changes on pull
• Rich Metadata — Categories, tags, thumbnails, dates, locations, and more
• Safe Sync — Detects conflicts when multiple users edit the same table
• Claude Code — CLI skill and MCP for AI-assisted workflows

Quick Start

# No install required
npx @memoreru-sdk/cli --help

# Or install globally
npm install -g @memoreru-sdk/cli

mem is a shorthand alias for memoreru. Both work identically.

Setup

Option A: CLI Login (Recommended)

memoreru login

Opens your browser for authentication. After login, create an API key:

memoreru keys create

API keys require the Light plan or above. A 14-day free trial is available.

Option B: Manual API Key

1. Log in to Memoreru
2. Go to Settings > Security > API Keys
3. Create a key with Read + Write access

export MEMORERU_API_KEY=your-api-key

That's it! Now you can use memoreru init, memoreru pull and memoreru push.

Usage

Login / Logout

memoreru login                        # Browser login
memoreru login --profile work         # Save as named profile
memoreru logout                       # Clear default profile
memoreru logout --all                 # Clear all profiles

Keys

memoreru keys create                          # Create API key (read + write)
memoreru keys create --name "my-key"          # With custom name
memoreru keys create --read-only              # Read-only key
memoreru keys list                            # List all keys
memoreru keys revoke mk_xxxxx                 # Revoke by prefix

Init

memoreru init ./my-page                  # Page (default)
memoreru init ./my-table --type table    # Table
memoreru init ./my-folder --type folder  # Folder

Pull

memoreru pull                    # Pull to current directory
memoreru pull ./my-data          # Pull to specific directory
memoreru pull ./my-data --preview  # Preview without writing

Supported types:

• Folder → directory structure
• Page / Slide → .md (images → images/)
• Table → .csv
• View / Graph / Dashboard → .json

View / Graph / Dashboard settings contain internal IDs. Pull/push works within the same environment but not across different environments.

Push

memoreru push                    # Push from current directory
memoreru push ./my-data          # Push from specific directory
memoreru push ./my-data --preview  # Preview without uploading

Status

Show local changes since the last pull or push.

memoreru status                  # Show changes in current directory
memoreru status ./my-data        # Show changes in specific directory

  memoreru status

  Modified:
    M  readme.md (page) "Project README" [body]

  New (not yet pushed):
    +  new-page.md (page) "New Page"

  2 content(s): 1 modified, 1 new

No API key required — works entirely offline.

Diff

Show unified diffs for modified files.

memoreru diff                          # Show all diffs
memoreru diff --file readme.md         # Show diff for a specific file

diff --git a/readme.md b/readme.md
--- a/readme.md
+++ b/readme.md
@@ -3,7 +3,7 @@
 ## Section 1

-Old content
+New content

 ## Section 2

No API key required — compares against locally stored snapshots. Output is git apply compatible.

File Structure

The CLI uses .memoreru.json manifests. Only listed items are synced.

my-project/
├── .memoreru.json          # Manifest
├── .memoreru/              # Sync state (auto-generated, add to .gitignore)
├── readme.md               # Page
├── tasks.csv             # Table (with row_id + version after push)
├── tasks.bak.csv         # Original CSV backup (auto-generated on first push)
├── docs/                   # Folder
│   ├── .memoreru.json      # Manifest for docs/
│   └── guide.md
└── images/
    └── logo.png            # Referenced from Markdown

Keys are file or directory names only. Folder contents are not auto-uploaded — place a .memoreru.json in each subdirectory.

Add .memoreru/ to your .gitignore — it stores sync snapshots for status and diff, not source content.

.memoreru.json

{
  "readme.md": {
    "content_type": "page",
    "title": "Project README"
  },
  "tasks.csv": {
    "content_type": "table",
    "title": "Tasks"
  },
  "docs": {
    "content_type": "folder",
    "title": "Documentation"
  }
}

After the first push, IDs are automatically written back:

{
  "readme.md": {
    "content_id": "q589jor87vmbnyylb8091cik",
    "content_type": "page",
    "title": "Project README"
  },
  "tasks.csv": {
    "content_id": "dyn8dapi7ckz8vvic8indjnc",
    "content_type": "table",
    "title": "Tasks",
    "columns": [
      { "id": "col_abc123", "name": "Title", "type": "string" },
      { "id": "col_def456", "name": "Status", "type": "string" },
      { "id": "col_ghi789", "name": "Priority", "type": "string" }
    ]
  }
}

For tables, columns maps each column name to its unique server-side ID. This ensures views and graphs referencing these columns survive table re-creation.

Properties

Only content_type is required. All other properties are optional.

Basic Info

Property	Description
content_id	Auto-set after first push
content_type	Required. folder, page, table, slide, view, graph, dashboard
title	Inferred from filename if omitted
scope	public, team, private (default: private)
description	Description text
description_expanded	Show expanded (default: false)
category	Category name or key
label	Label
tags	Tag names (e.g., ["React", "Tutorial"])
slug	Custom URL slug (paid plan)

Additional Info

Property	Description
thumbnail	Image path (e.g., ./images/thumb.png)
icon	Icon (emoji or image): {"type":"emoji","emoji":"📝"} / {"type":"image","path":"./icon.png"} / null
datetime	Date/time object: {"start":"2026-01-15","end":null,"type":"date"}
location	Location object: {"area1":null,"area2":null,"lat":35.6,"lng":139.7,"address":"...","name":"..."}
persons	Person names (e.g., ["John Doe"])
sources	References
language	Language code (default: en)
extensions	Extensions (style/script/custom_process) by file path (see below)

Publishing

Property	Description
team_id	Team ID (required when scope is team)
parent_content_id	Parent folder's content_id
publish_status	draft or published (default: published)
scheduled_at	Scheduled publish time (ISO 8601)
expires_at	Expiration time (ISO 8601)
is_suspended	Suspend (default: false)
is_archived	Archive (default: false)

Privacy

Property	Description
discovery	listed, unlisted, profile (default: listed)
access_level	open, login_required, followers_only (default: open)
can_embed	Allow embedding (default: true)
can_ai_crawl	Allow AI crawlers (default: true)
has_password	Password protection (default: false)

Other

Property	Description
is_pinned	Pin to top (default: false)
is_locked	Lock editing (default: false)
auto_summary	Auto-generate summary (default: false)
auto_translate	Auto-generate translations (default: false)

Table Columns (auto-managed)

Property	Description
columns	Column definitions (table only). Auto-set after push. Array of { id, name, type }.

Column IDs ensure unique references from views and graphs. To rename a column, update both the CSV header and columns[].name in .memoreru.json.

Table CSV

Before first push (original CSV):

Title,Status,Priority
Setup CI,Done,High
Write docs,In Progress,Medium

After first push (row_id + version added, original backed up as .bak.csv):

row_id,version,Title,Status,Priority
row_abc123,1,Setup CI,Done,High
row_def456,1,Write docs,In Progress,Medium

• row_id and version are system-managed columns (first two columns)
• Column types are auto-inferred on push
• Only changed rows are sent on subsequent pushes (diff-based)
• Version mismatch triggers conflict detection for collaborative editing

Adding a new row — leave row_id and version empty:

row_id,version,Title,Status,Priority
row_abc123,1,Setup CI,Done,High
row_def456,1,Write docs,In Progress,Medium
,,Add tests,Todo,Low

Images

![Screenshot](./images/screenshot.png)

• Pull: Download only changes
• Push: Upload automatically

Options

--api-key <key>     API key (overrides MEMORERU_API_KEY)
--profile <name>    Credential profile name
--url <url>         Base URL (default: https://memoreru.com)
--help              Show help
--version           Show version

Environment variables:

Variable	Description
MEMORERU_API_KEY	API key
MEMORERU_TENANT	Tenant slug for verification (recommended for dedicated tenants)
MEMORERU_URL	Base URL (default: https://memoreru.com)

Dedicated tenant users: Set MEMORERU_TENANT to prevent accidental pushes to the wrong tenant. The CLI verifies the tenant before each pull/push and stops if it doesn't match.

Auth Resolution Order

1. --api-key flag
2. MEMORERU_API_KEY env var
3. --profile flag
4. .memoreru-config.json profile
5. Default profile in ~/.config/memoreru/credentials.json

.memoreru-config.json

Per-directory profile config for multi-account workflows:

{
  "profile": "work"
}

When running memoreru push ./my-dir, the CLI reads .memoreru-config.json from that directory to determine which profile (and API key) to use.

Claude Code Integration

Combine with Claude Code to manage content using natural language.

CLI Skill (File Sync) — Teach Claude Code how to use pull/push:

cp -r node_modules/@memoreru-sdk/cli/skills/memoreru-cli ~/.claude/skills/

MCP (Data Operations) — Add .mcp.json to your project root:

{
  "mcpServers": {
    "memoreru": {
      "url": "https://memoreru.com/api/mcp/",
      "headers": {
        "Authorization": "Bearer your-api-key"
      }
    }
  }
}

When to use which:

• CLI — File-based sync with Git version control
• MCP — Direct data operations (add rows, search, update)

Programmatic API

import { configure, pullContent, pullTableData, upsertContent } from '@memoreru-sdk/cli';
import type { UpsertInput } from '@memoreru-sdk/cli';

configure({
  baseUrl: 'https://memoreru.com',
  apiKey: 'your-api-key',
});

const page = await pullContent('q589jor87vmbnyylb8091cik');
const table = await pullTableData('dyn8dapi7ckz8vvic8indjnc');

const input: UpsertInput = {
  content_type: 'page',
  title: 'My Page',
  body: '# Hello',
};
const result = await upsertContent(input);

License

MIT

Trademark

"Memoreru" is a registered trademark in Japan (Class 9, Class 42 / Registration No. 7037589).

---

Made with for knowledge creators

Sync your content, own your workflow!