Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@journalowl-mcp Create a journal entry about my productive day"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP (Model Context Protocol) server for JournalOwl - AI-powered journaling integration for Claude Code, Cursor, and ChatGPT Agents.
Create journal entries directly from your AI assistant
Search and browse your journal entries
Access weekly reviews with emotional trends and insights
Get personalized suggestions based on your writing style
Secure API key authentication with scoped permissions
Related MCP server: ai-journal
npx @mindfulabai/journalowl-mcpOr install globally:
npm install -g @mindfulabai/journalowl-mcpLog in to JournalOwl
Go to Settings > API Keys
Create a new API key with the scopes you need:
journal:read - Read your journal entries
journal:write - Create new entries
review:read - Access weekly reviews
profile:read - Access your profile and writing style
Add to your ~/.claude.json:
{
"mcpServers": {
"journalowl": {
"command": "npx",
"args": ["@mindfulabai/journalowl-mcp"],
"env": {
"JOURNALOWL_API_KEY": "jowl_sk_your_api_key_here"
}
}
}
}Add to your Cursor MCP settings:
{
"journalowl": {
"command": "npx",
"args": ["@mindfulabai/journalowl-mcp"],
"env": {
"JOURNALOWL_API_KEY": "jowl_sk_your_api_key_here"
}
}
}Tool | Description |
| Create a new journal entry |
| List entries with filters |
| Get entry details and analysis |
| Search entries by text |
| Get weekly review and insights |
| Get writing style preferences |
URI | Description |
| User profile and journaling stats |
| Recent entries metadata |
Once configured, you can use JournalOwl directly in your AI conversations:
"Create a journal entry about my productive day at work"
"Show me my journal entries from last week"
"What insights does my weekly review show?"
"Search my journal for entries about anxiety"# Clone the repository
git clone https://github.com/mindfulabai/journalowl-mcp.git
cd journalowl-mcp
# Install dependencies
npm install
# Build
npm run build
# Run in development mode
npm run dev
# Test with MCP Inspector
npm run inspectorAPI keys are transmitted securely via HTTPS
Keys are hashed in our database (we never store the plain key)
You can revoke keys at any time from JournalOwl settings
Scopes limit what each key can access
MIT
Create a new journal entry in JournalOwl. Entry is created with status "in_progress". Use journal_finalize_entry to generate AI analysis and complete the entry.
Inputs: content (required), date, mood, tags.
Finalize a journal entry and generate AI analysis. This generates sentiment analysis, themes, and insights. Entry must have at least 100 characters.
Inputs: entry_id (required).
Get a specific journal entry by ID, including its AI analysis and insights.
Inputs: entry_id (required), include_analysis.
Get a weekly review summarizing journaling activity, emotional trends, and insights. Use "latest" to get the most recent review.
Inputs: week.
Get the user's preferred writing style, tone, and personalized journaling suggestions.
Inputs: no named parameters.
List journal entries from JournalOwl. Use filters to find specific entries by date, status, or tags.
Inputs: tags, limit, offset, status, +2 more.
Search journal entries by text. Find entries related to specific topics, emotions, or events.
Inputs: query (required), limit.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 7 tools
Each tool targets a distinct action, but journal_list_entries and journal_search both retrieve entries with different filter types (metadata vs text), which could cause minor confusion. Overall the purposes are clear.
All tools share the journal_ prefix and use snake_case, but journal_search is verb-only while others follow verb_noun (e.g., journal_get_entry), creating a slight inconsistency.
Seven tools is well-scoped for a journaling server, covering creation, retrieval, search, and analysis without unnecessary redundancy or bloat.
The create-read workflow is well-covered (create, finalize, list, get, search), but there are no update or delete tools for entries, and no way to modify writing style preferences, creating notable lifecycle gaps.
One MCP server for the services your AI needs: memory, files, Microsoft 365, kanban and more.
MCP server for AI dialogue using various LLM models via AceDataCloud
Markdown-based note-taking with a hosted MCP server. Your notes serve you and your AI.
Hosted MCP server connecting AI assistants to 9,000+ apps and 40,000+ actions via Zapier.
