sandboxed-execution

Local-first, self-hosted AI agent runtime and MCP bridge with sandboxed sessions, memory, credentials, audit/replay, and a local Console.

Secure MySQL Administration & Observability Featuring Code Mode— One Tool Replacing All 242 Specialized Tools for up to 90% Token Savings. Includes Connection Pooling, HTTP/SSE, OAuth 2.1, Deterministic Error Handling, Advanced Encryption, Integrated Observability, and Full Support for ProxySQL, MySQL Router & MySQL Shell.

🧬 RAGIX: Local-first development assistant making LLMs behave like disciplined engineers – Unix-RAG retrieval, sandboxed execution, MCP-compatible, fully auditable

Local AI Workstation CLI for LM Studio/Ollama. SysAdmin, NetAdmin, and Senior Dev in your terminal. Maps codebases, previews edits, autonomous build recovery. Deep-kernel health, crash logs, and network stack. 54-voice TTS + Hybrid BM25/Semantic search with PageRank graph and heat-weighted memory. Windows-first. One Rust binary. No cloud. No API.

langgraph langchain llm ai-agent python fastapi streamlit groq gemini openai human-in-the-loop multi-agent code-generation langsmith celery redis

Parallel OpenAI Codex agents inside Claude Code — multi-agent orchestration with git worktree isolation, resumable sessions, and dual-model PR review. MCP plugin.

Native companion for OS Loop AI that extends browser agents with secure local capabilities—filesystem access, MCP connectivity, long-running tasks, and system-level execution

A portable, offline coding agent that runs entirely from a USB drive on Windows. It uses a local LLM through llama.cpp for GPU-accelerated inference, so no cloud APIs or internet connection are needed once set up. Includes sandboxed file tools, a simple API for generating coding projects, and packages finished work into downloadable ZIP artifacts.

Self-hosted gateway for running LLM-generated Python and Bash in isolated Docker sandboxes, with PPTX and LaTeX rendering.

🖥️ A self-hosted MCP control plane for isolated ChatGPT coding sandboxes

AX hackathon project workspace for designing, implementing, and validating multi-agent AI solutions.

Autonomous research agents that read the literature, weigh the evidence, and run the experiment — then hand you a cited, reproducible answer.

AI-powered detective agents for log analysis and debugging using Runloop's ephemeral infrastructure. Perfect for ad-hoc investigations without permanent setup.

CodeForge — AI Code Agent with Self-Repair. Full-stack FastAPI + Next.js app with multi-agent orchestration, sandboxed code execution, and LLM routing.

A high-performance, cross-language Meta-Compiler and VM environment built with Rust. It provides a unified runtime to execute Python, C++, Rust, and other languages through a single binary, utilizing LLVM IR for seamless interoperability and zero-config deployment

Local AI Workstation CLI 2026 – SysAdmin, NetAdmin & Senior Dev in Your Terminal

AI Code Agent Workbench 2026 - Multi-Model Orchestration & Git Isolation

Orchestrate intelligent multi-agent swarms that actually work. Deploy autonomous workflows and conversational AI systems natively in Node.js. Features an enterprise-grade architecture, self-learning swarm intelligence, dynamic context compression, and persistent RAG memory without the brittle CLI wrappers or enterprise framework bloat.

Think. Review. Code. Safely.

Local-first autonomous AI agent for Mobile & Desktop. Gemma 4 / Ollama offline + waterfall failover to 7 cloud APIs. Flutter/Dart — premium ANSI TUI, Vim-modal input, 41+ tools (MCP, shell, sub-agents, LSP, Git), multi-agent swarm, self-healing loops. 40K+ LOC. Zero cost. Full privacy.