Code Knowledge Tool

A knowledge management tool for code repositories using vector embeddings. This tool helps maintain and query knowledge about your codebase using advanced embedding techniques.

Building and Installing

1. Build the Package

First, you need to build the distribution files:

This will create two files in the dist/ directory:

code_knowledge_tool-0.1.0-py3-none-any.whl (wheel file for installation)

code_knowledge_tool-0.1.0.tar.gz (source distribution)

2. Install the Package

Prerequisites

Ensure Ollama is installed and running:

Install the package:

Option 1: Install from wheel file (recommended for usage)

Option 2: Install in editable mode (recommended for development)

This option is best if you want to modify the tool or contribute to its development:

Integration with RooCode/Cline

Copy the MCP configuration to your settings:

For Cline (VSCode):

Add this configuration:

For RooCode:

Add the same configuration as above.

Restart RooCode/Cline to load the new tool.

Using as Memory Bank and RAG Context Provider

This tool can serve as your project's memory bank and RAG context provider. To set this up:

Copy the provided template to your project:

Customize the rules and patterns in .clinerules for your project's needs

The template includes comprehensive instructions for:

Knowledge base management

RAG-based development workflows

Code quality guidelines

Memory management practices

See clinerules_template.md for the full configuration and usage details.

Features

Local vector storage for code knowledge

Efficient embedding generation using Ollama

Support for multiple file types

Context-aware code understanding

Integration with RooCode and Cline via MCP

RAG-based context augmentation

Persistent knowledge storage

Requirements

Python 3.8 or higher

Ollama service running locally

chromadb for vector operations

Development

Running Tests

The project follows an integration-first testing approach, focusing on end-to-end functionality and MCP contract compliance. The test suite consists of:

MCP Contract Tests - Tool registration and execution - Resource management - Knowledge operations - Error handling

Package Build Tests - Installation verification - Dependency resolution - MCP server initialization - Basic functionality

To run the tests:

Test Environment Requirements:

The tests use a temporary directory (test_knowledge_store) that is cleaned up automatically between test runs.

For more details on the testing strategy and patterns, see the documentation in `docs/`.

Future Distribution

If you want to make this package available through pip (i.e., `pip install code-knowledge-tool`), you would need to:

Install twine: `pip install twine`

Upload your distribution: `twine upload dist/*`

However, for now, use the local build and installation methods described above.

License

MIT License

Code Embeddings