Skip to content

cloudcover95/JuniorPython-Suite

Repository files navigation

JuniorPython Suite Modular Python IDE & Edge-Compute SDKJuniorPython is a logic-dense, power-efficient development suite engineered for local-first execution on Apple Silicon (M4/M1) and Windows 11 (CUDA) environments. It operates as a sovereign manifold, bypassing cloud reliance for LLM inference, data processing, and automation.Core ArchitectureDynamic Plugin System: Tools are isolated .py scripts located in /tools/, hot-reloaded via importlib.Data Pipeline: High-throughput ingestion using Snappy-compressed .parquet vectors for zero-copy memory access.Multiprocess Sandbox: Isolated execution kernel for safe script testing with hardware resource monitoring (RAM/CPU).Security Protocol IV: Explicit path isolation for 01_Legal and 02_Assets to prevent agent traversal or indexing.Deployment ProtocolInitialize Hardware: Ensure Ollama is running locally for neural inference.Provision Environment:pip install -r requirements.txt ollama pull llama3.2:3b Execute Matrix:python main.py Repository Topologycore/: System kernels (Sandbox, Pipeline, BaseTool).tools/: Extensible tool manifold (Branches 01-27)..jp_cache/: Localized Parquet/JSONL data vault.

PyForge SuiteModular Python IDE and Automation Powerhouse for Windows 11.Engineered for CUDA-accelerated video rough cuts and local LLM agentic workflows.Deployment ProtocolInitialize environment:python -m venv venv call venv\Scripts\activate pip install -r requirements.txt Ensure Ollama is running locally:ollama serve ollama pull llama3.2 Execute core runtime:python main.py

About

Suite architecture is designed for local-first, low-latency execution, utilizing dynamic module loading (importlib) to treat its own ecosystem as a mutable runtime environment. The system leverages PySide6 for the GUI, sandboxing tools into an extensible registry. The llm_agent handles autonomous tool generation and hardware-accelerated via CUDA

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages