Personal collection of Python utilities following clean code principles with modular architecture and shared toolkit foundation.

Transform Unreal Engine 5 blueprint .COPY files into AI-friendly summaries.

• Purpose: Parse complex blueprint exports into markdown/JSON
• Features: Variable extraction, graph analysis, function call tracking
• Tech: Python 3.13, uv, custom parsers
• New to Blueprints? See the Blueprint .COPY guide to understand what the parser reads.

→ Documentation

HTTP API wrapper for Ollama providing LLM query capabilities with FastMCP integration.

• Purpose: Expose Ollama API via REST endpoints
• Features: GET/POST endpoints, structured logging, health checks
• Tech: FastAPI, httpx, FastMCP, structlog

→ Documentation

Shared utilities for text parsing, file I/O, formatting, and console output.

• Purpose: Reusable helpers across all projects
• Modules: text_parsing, file_io, collections, formatting, console
• Tech: Pure Python, lightweight

→ Documentation

All projects leverage a common toolkit package:

from toolkit.text_parsing import find_first_match, extract_tail_identifier
from toolkit.file_io import read_text_with_encoding_detection
from toolkit.console import print_metric_summary

Every project follows strict guidelines:

• Functions < 20 lines (ideally 5-10)
• Single responsibility per function
• One abstraction level per function
• Comprehensive Doxygen documentation
• Stepdown rule for readability

project/
├── src/project_name/
│   ├── parsers/          # Data extraction
│   ├── formatters/       # Output rendering
│   ├── models.py         # Data structures
│   ├── pipeline.py       # Orchestration
│   └── ...
├── data/
│   ├── in/               # Input files
│   └── out/              # Generated output
├── tests/                # Test scaffolds (not enforced yet)
├── README.md
└── pyproject.toml

# Clone or navigate to Utils folder
cd Utils

# Each project uses uv for dependency management
cd blueprint-cleaner
uv sync
uv run blueprint-cleaner --help

cd ../ollama-service
uv sync
uv run uvicorn ollama_service.api:app --reload --port 4001

Projects declare local dependencies via tool.uv.sources:

[tool.uv.sources]
toolkit = { path = "../", editable = true }
ollama-service = { workspace = true, editable = true }

Each package includes comprehensive documentation:

• Individual project READMEs with Quick Start guides
• API documentation and usage examples
• Architecture and design decisions

• Python: 3.13+
• Package Manager: uv 0.9.2
• Web Framework: FastAPI
• MCP Integration: FastMCP
• Build Backend: hatchling / uv_build

1. Create new utility:

   uv init --package new-project
   cd new-project
   # Add toolkit as dependency in pyproject.toml

2. Implement with clean code principles:

   • Small functions (5-10 lines)
   • Doxygen comments
   • Modular structure
   • Single responsibility

3. Test scaffolding:

   mkdir tests
   # Add placeholder tests

4. Documentation:

   • README with Quick Start
   • Architecture section
   • API examples
   • Module documentation

Tests are scaffolded but intentionally not enforced:

• Structure ready for pytest
• Placeholder tests prevent import errors
• Run tests: uv run pytest

MIT