Codebase-X: Production-grade, offline code comprehension and AI-tasking platform
Project description
Codebase-X
The Intelligent AI Assistant for Your Entire Codebase.
Codebase-X is a production-grade, offline platform designed to give you total control over your code. By building a deep, persistent Knowledge Graph of your repository, Codebase-X enables advanced architectural analysis, atomic task decomposition, and prompt engineering—all while keeping your data 100% private and local.
🗺️ User Journey
codebookx analyze: Scan your repository to build a Knowledge Graph and architectural teardown.codebookx view: Launch the interactive Knowledge Graph browser to explore your code visually.codebookx ask "...": Ask technical questions about your codebase, powered by the Knowledge Graph context.
🚀 Core Capabilities
🏛️ Codebase Teardown (Analyze)
Stop guessing how your system works. Codebase-X scans your entire repository to generate a high-level architectural blueprint.
- Elevator Pitch: Get an instant summary of any project.
- Architecture Blueprint: Automatically mapped folder roles and purposes.
- Data Flow: Understand how the core components of your system interact.
codebookx analyze
🔨 Atomic Task Decomposition (Decompose)
Break complex features into small, manageable pieces. Codebase-X analyzes your feature request against your actual codebase to produce a series of atomic tasks.
- AI-Ready: Each task includes a "copy-paste" prompt specifically designed for AI coding assistants.
- Context-Aware: Tasks reference the exact files and symbols needed.
codebookx decompose "Add a user settings page"
✨ Prompt Enhancement (Enhance)
Stop writing vague prompts. The Prompt Enhancer uses the Knowledge Graph to inject relevant file paths, function signatures, and dependencies into your request, creating a high-context prompt that gets it right the first time.
codebookx enhance "Refactor the authentication flow"
🕸️ Knowledge Graph Browser (View)
Explore your code visually. Launch an interactive local web UI to traverse your project's dependency graph, find symbols, and manage your AI tasks.
- Visual Map: Interactive D3.js force-directed graph available at
/graph. - Relations: Track CALLS, IMPORTS, and CONTAINS relationships cross-file.
codebookx view
❓ Technical Q&A (Ask)
Knowledge Graph-backed Q&A for your codebase. Answers are saved automatically.
# Single question
codebookx ask "How is the authentication handled in this project?"
# Interactive chat mode (multi-turn, context carry-over)
codebookx ask -c "Walk me through the data flow"
# Save Q&A logs to a custom directory
codebookx ask "What does run_ask do?" --dir ~/my-notes
- Auto-save: Every Q&A and chat session is saved to
ask_history/by default (--dirto override, or setCODEBOOK_ASK_DIRenv var). - Chat mode (
-c): Multi-turn conversation with full context carry-over. Each turn is saved incrementally — no history lost on crash.
🛠️ Getting Started
1. Install
pip install .
2. Local LLM Setup Codebase-X is designed to work with LM Studio, Ollama, or any OpenAI-compatible local server. Ensure your server is running.
3. Initial Analysis Scan your repository to build the Knowledge Graph:
codebookx analyze
🔒 Privacy & Performance
- 100% Offline: Your code never leaves your machine. No cloud APIs, no subscriptions.
- Persistent Indexing: Uses a lightning-fast SQLite-backed Knowledge Graph with hash-based caching.
- Multi-Language: Deep support for Python, TypeScript, JavaScript, and more.
📄 License
Codebase-X is released under the MIT License.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file codebookx-3.0.1.tar.gz.
File metadata
- Download URL: codebookx-3.0.1.tar.gz
- Upload date:
- Size: 23.8 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.1
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
20327c2d2fcf2d537cf29ab19bd690ad45da2ce66b80d412399ea0b356015516
|
|
| MD5 |
1a618c8e7bbbefe50eb849d1fce27912
|
|
| BLAKE2b-256 |
5fcbd138f014c7dae8aa4c97730d495549d5eebe8a30e8632b0528449126f2e0
|
File details
Details for the file codebookx-3.0.1-py3-none-any.whl.
File metadata
- Download URL: codebookx-3.0.1-py3-none-any.whl
- Upload date:
- Size: 25.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.1
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
c7e46d0106d617c2c053d0715b67617651abe33996aa2d1077bad4e30c12d493
|
|
| MD5 |
8d5c29f0599524e5b1d64aa81d45ba48
|
|
| BLAKE2b-256 |
232517b7e92b2ff882a2ef8d835df7c9acc55c7b0405defb7eea5d5b47b9c160
|