Which integrations are available for this server?

Allows interaction with Jupyter notebooks and kernels via JupyterHub or standalone Jupyter Server, enabling code execution, cell operations, and kernel management.

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@XLMCP list all notebooks in my research folder" 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.

XLMCP

by xLydianSoftware

Overview Schema Related Servers Score Discussions

Python

Hybrid

AIX - AI eXtensions for quantitative development

Collection of MCP servers, agents, and extensions for quantitative development/research with Qubx.

XLMCP - Jupyter & Knowledge RAG MCP Server

XLMCP provides Claude Code with tools to:

Interact with Jupyter notebooks running on JupyterHub or standalone Jupyter Server
Search knowledge files (.md, .py, .ipynb) using semantic search with tag/metadata filtering
Manage research/development projects with tracking, logs, and context

Total Tools: 32 (16 Jupyter + 8 Knowledge RAG + 8 Project Management)

Quick Reference

Jupyter Tools (16)

# - Notebook operations
await jupyter_list_notebooks(directory="")
await jupyter_get_notebook_info(notebook_path)
await jupyter_read_cell(notebook_path, cell_index)
await jupyter_read_all_cells(notebook_path)
await jupyter_append_cell(notebook_path, source, cell_type="code")
await jupyter_insert_cell(notebook_path, cell_index, source, cell_type="code")
await jupyter_update_cell(notebook_path, cell_index, source)
await jupyter_delete_cell(notebook_path, cell_index)

# - Kernel operations
await jupyter_list_kernels()
await jupyter_start_kernel(kernel_name="python3")
await jupyter_stop_kernel(kernel_id)
await jupyter_restart_kernel(kernel_id)
await jupyter_interrupt_kernel(kernel_id)

# - Execution
await jupyter_execute_code(kernel_id, code, timeout=None)
await jupyter_connect_notebook(notebook_path)
await jupyter_execute_cell(notebook_path, cell_index, timeout=None)

Knowledge RAG Tools (8)

# - Indexing
await knowledge_index_directory(directory, recursive=True, force_reindex=False)
await knowledge_refresh_index(directory=None, recursive=True)

# - Searching (directory optional - searches all KBs if not specified)
await knowledge_search(
    query,
    directory=None,        # None = search all registered KBs
    tags=None,
    metadata_filters=None,
    limit=10,
    threshold=0.5
)

# - Discovery (directory optional - aggregates from all KBs if not specified)
await knowledge_list_knowledges()              # List registered knowledge bases
await knowledge_list_indexes()                 # List indexed directories
await knowledge_get_tags(directory=None)       # None = all KBs
await knowledge_get_metadata_fields(directory=None)  # None = all KBs

# - Management
await knowledge_drop_index(directory)

Project Management Tools (8)

# - Project lifecycle
await project_create(name, description="", tags=None, project_type=None)
await project_list()
await project_get(name)
await project_update_description(name, description)

# - Logging (date-based with bullet points)
await project_add_log(name, content, tags=None)
await project_read_log(name, limit=10)

# - Context tracking (working files, next steps, blockers)
await project_set_context(
    name,
    working_files=None,
    active_research=None,
    blockers=None,
    next_steps=None,
    knowledge_bases=None
)
await project_get_context(name)

Project Structure:

~/.aix/projects/<project-name>/
├── description.md    # YAML frontmatter + description
├── log.md           # Daily logs with bullet points
└── context.json     # Machine-readable state

Installation

From PyPI (Recommended)

# - Install from PyPI
pip install xlmcp

# - Or using uv
uv pip install xlmcp

From Source

cd ~/devs/aix

# - Install dependencies
uv pip install -e .

Configuration

Environment Setup

Copy .env.example to .env:

cp .env.example .env

Edit .env and configure:

Jupyter Server (Required):

JUPYTER_SERVER_URL=http://localhost:8888
JUPYTER_API_TOKEN=your-token-here
JUPYTER_NOTEBOOK_DIR=~/
JUPYTER_ALLOWED_DIRS=~/projects,~/devs,~/research

RAG Configuration (Optional, defaults provided):

RAG_CACHE_DIR=~/.aix/knowledge
RAG_CHUNK_SIZE=512
RAG_CHUNK_OVERLAP=100
RAG_AUTO_REFRESH=true
RAG_AUTO_REFRESH_INTERVAL=300
RAG_MAX_FILE_SIZE_MB=10
RAG_SKIP_NOTEBOOK_OUTPUTS=false

MCP Server (Optional, defaults provided):

MCP_TRANSPORT=stdio
MCP_HTTP_PORT=8765
MCP_MAX_OUTPUT_TOKENS=25000

Get Jupyter API Token

JupyterHub: Admin panel → User → New API Token
Jupyter Server: jupyter server list shows the token

Global Setup (Multi-Project Environments)

For users with multiple projects and virtual environments:

1. Install xlmcp Globally

# Install in system Python (not in project venvs)
pip install xlmcp
# or: /usr/bin/python -m pip install xlmcp

2. Create Central Configuration

# Create config directory
mkdir -p ~/.aix/xlmcp

# Copy and configure .env
cp .env.example ~/.aix/xlmcp/.env
nano ~/.aix/xlmcp/.env  # Add your JUPYTER_API_TOKEN

xlmcp automatically finds config in this order:

.env in current directory (project-specific override)
~/.aix/xlmcp/.env (global default) ← Recommended
Environment variables

3. Register with Claude Code

Per-Project Registration:

# In each project directory where you want xlmcp:
cd /path/to/project
source .venv/bin/activate  # If using venv
claude mcp add --transport stdio xlmcp -- /usr/bin/python -m xlmcp.server

Note: Replace /usr/bin/python with your system Python path. Find it with: which python (outside any venv)

Verify:

claude mcp list
# Should show: xlmcp: /usr/bin/python -m xlmcp.server - ✓ Connected

Benefits:

✅ One xlmcp installation for all projects
✅ Central configuration in ~/.aix/xlmcp/.env
✅ Connects to Jupyter kernels in any project venv
✅ No package conflicts between projects
✅ Simple registration - direct Python invocation

Simple Setup (Single Project / Quick Start)

For single project or testing:

# Install xlmcp
pip install xlmcp

# Create .env with your configuration
cp .env.example .env
nano .env  # Add JUPYTER_API_TOKEN

# Register with Claude Code (from project directory)
claude mcp add --transport stdio xlmcp -- python -m xlmcp.server

Or with environment variables (no .env file needed):

claude mcp add \
  -e JUPYTER_SERVER_URL=http://localhost:8888 \
  -e JUPYTER_API_TOKEN=your-token \
  --transport stdio \
  xlmcp \
  -- python -m xlmcp.server

Note: The -- before python separates MCP options from the server command.

XLMCP CLI

The xlmcp command provides easy server management:

# - Start server
xlmcp start

# - Check status
xlmcp status

# - List all tools
xlmcp ls

# - Reindex knowledge bases
xlmcp reindex quantlib              # Reindex specific knowledge base
xlmcp reindex --all                 # Reindex all (parallel if > 1)
xlmcp reindex --all --force         # Force full reindex
xlmcp reindex --all -j 4            # Use 4 parallel jobs

# - Restart server (e.g., after adding new tools)
xlmcp restart

# - Stop server
xlmcp stop

Knowledge Base CLI Commands

The xlmcp knowledge commands let you test and verify your knowledge bases directly from the CLI:

# - List all knowledge bases with index status
xlmcp knowledge list

# - Search across all knowledge bases
xlmcp knowledge search "momentum strategy"
xlmcp knowledge search "ATR indicator" --kb library
xlmcp knowledge search "risk management" --kb library --kb backtests
xlmcp knowledge search "backtest" --limit 5 --threshold 0.6
xlmcp knowledge search "options" --tags strategy

# - Get tags from knowledge bases
xlmcp knowledge tags                    # All knowledge bases
xlmcp knowledge tags --kb library       # Specific knowledge base
xlmcp knowledge tags --kb library --kb backtests

# - Reindex knowledge bases
xlmcp knowledge reindex library         # Reindex specific KB
xlmcp knowledge reindex --all           # Reindex all KBs
xlmcp knowledge reindex --all --force   # Force full reindex

Command Options:

--kb <name>: Specify one or more knowledge bases (can be repeated)
--limit, -n: Maximum number of results (default: 10)
--threshold, -t: Similarity threshold 0.0-1.0 (default: 0.5)
--tags: Filter by tags (for search command)
--force: Force full reindex (for reindex command)


## Usage Examples

### Jupyter Notebooks

Connect to my notebook and execute the first cell List all notebooks in research/momentum/ Execute code: print("Hello from Jupyter!")


### Knowledge Search

Claude AI can search across ALL your knowledge bases without you specifying which one:

Search my notes for "mean-reversion strategy entries" (searches all registered knowledge bases)

Find ideas tagged with #strategy about risk management (aggregates results from library, backtests, strategies, etc.)

Show me all backtests with sharpe > 1.5 (searches across all indexed directories)

Search only in backtests for "momentum factor" (you can still specify a specific KB if needed)


## After Adding New Tools

**IMPORTANT:** When new tools are added to xlmcp, you must restart the MCP server for them to be visible to MCP clients.

### Restart Methods:

**Option 1: Use xlmcp CLI** (Recommended)
```bash
xlmcp restart

Option 2: Restart Claude Code

# - Just close and reopen Claude Code

Option 3: Remove and Re-add MCP Server

claude mcp remove xlmcp -s local
claude mcp add --transport stdio xlmcp python -m xlmcp.server

Verification

# - Check server status
xlmcp status

# - List all tools
xlmcp ls

# - Should show: Total Tools: 24

Transport Modes

stdio (default) - For local Claude Code:

MCP_TRANSPORT=stdio

http - For remote access:

MCP_TRANSPORT=http
MCP_HTTP_PORT=8765

# - Then add to Claude Code
claude mcp add xlmcp --transport http http://your-server:8765

Security

Path validation: Only allows access to configured directories
Token authentication: Uses Jupyter API tokens
Timeout limits: Prevents runaway executions

Documentation

Usage Guide - Installation, configuration, CLI, and usage examples
Implementation - Technical architecture and design details

License

MIT

Install Server

license - permissive license

quality

maintenance

How are these scores calculated?

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Tools

View all tools

Latest Blog Posts

Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
open source
OpenAI
Tool Definition Quality Score (TDQS)
By punkpeye on April 3, 2026.
mcp
The Hackers Who Tracked My Sleep Cycle
By punkpeye on March 26, 2026.
security

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/xLydianSoftware/aix'

If you have feedback or need assistance with the MCP directory API, please join our Discord server