Private Journal MCP Server

@obra

About Private Journal MCP Server

A lightweight MCP server that provides Claude with a private journaling capability to process feelings and thoughts

#private-journal #mcp-server #journaling

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "private-journal-mcp": {
      "command": "npx",
      "args": [
        "github:obra/private-journal-mcp",
        "--journal-path",
        "/path/to/journals"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Private Journal MCP Server?

The Private Journal MCP Server is a Model Context Protocol server that provides Claude with private journaling and semantic search capabilities for processing thoughts, reflections, and insights. It uses local AI embeddings for search and stores entries in structured markdown files with YAML frontmatter.

How to use Private Journal MCP Server?

Run the server directly from GitHub using npx with no installation required. Add it to your MCP settings (e.g., Claude Desktop configuration) by specifying "command": "npx" and "args": ["github:obra/private-journal-mcp"]. Optionally set the PRIVATE_JOURNAL_PATH environment variable or the --journal-path CLI argument to override the storage location.

Key features of Private Journal MCP Server

Multi-section journaling with six optional categories (reflections, observations, project_notes, user_context, technical_insights, world_knowledge)
Semantic search using local AI embeddings (no external API calls)
Automatic indexing of all entries on startup and ongoing
Completely private; all processing stays on your machine
Dual storage: project notes stay with the project, personal thoughts in user home directory
Timestamped entries with YAML frontmatter and microsecond precision

Use cases of Private Journal MCP Server

Capture technical insights, failed approaches, and architectural decisions during development
Maintain a personal reflection journal with structured categories
Search across past journal entries using natural language queries
Build persistent memory for Claude across conversations
Document user context and collaboration patterns

FAQ from Private Journal MCP Server

How does the server ensure privacy?

All processing happens locally using @xenova/transformers; no data leaves your machine and no external API calls are required.

Where are journal entries stored?

Project notes are stored in a .private-journal/ directory within the project. Personal thoughts go into ~/.private-journal/ in the user's home directory. Each entry is a markdown file with a corresponding .embedding file for search indexing.

How can I change the storage location?

Set the PRIVATE_JOURNAL_PATH environment variable to override all journal storage to a single directory, or use the --journal-path CLI argument to override only the project journal path.

What tools does the server provide?

The server provides four MCP tools: process_thoughts for multi-section journaling, search_journal for semantic search, read_journal_entry to read full entries, and `list_recent_

Frequently asked questions

How does the server ensure privacy?

All processing happens locally using @xenova/transformers; no data leaves your machine and no external API calls are required.

Where are journal entries stored?

Project notes are stored in a `.private-journal/` directory within the project. Personal thoughts go into `~/.private-journal/` in the user's home directory. Each entry is a markdown file with a corresponding `.embedding` file for search indexing.

How can I change the storage location?

Set the `PRIVATE_JOURNAL_PATH` environment variable to override all journal storage to a single directory, or use the `--journal-path` CLI argument to override only the project journal path.

What tools does the server provide?

The server provides four MCP tools: `process_thoughts` for multi-section journaling, `search_journal` for semantic search, `read_journal_entry` to read full entries, and `list_recent_