A reference architecture demonstrating local-first LLM orchestration for SOC compliance workflows. Built on ~11GB VRAM edge hardware using Ollama, featuring a multi-agent coordination pattern with file-based handoffs, a two-lane integration model (human-initiated vs automated processing), and a HITL approval gate for resource-sensitive operations.

Claude-Code-style CLI for your own local LLM fleet. Multi-agent delegation, MCP, hooks, autopilot, 100+ tools (file/shell/web/security/free APIs) — all running on your GPU, no cloud calls.

55+ AI agent skills for sales, content, research, and fleet operations. Free lite editions. Full alpha at koino.capital

Provide Hermes agents with persistent semantic memory using SQLite truth storage and LanceDB vector retrieval for cross-session context management.