Fixing Claude Code's amnesia with persistent memory
Scatter-gather retrieval with prompt caching beats naive vector search for Claude Code memory.
A governed runtime for persistent human-AI collaboration
Ambitious architecture: persistent hyperdimensional memory (QAIS), deterministic paragraph retrieval (SLA), and a dual-operator Save Protocol that forces human+AI agreement before writes. The practical touches — a React control panel, an AutoHotkey clipboard bridge/counter, and one-command installer — show someone built this to be used, not just theorized. Platform lock (Windows + AHK) and a README heavy on terminology mean it's exciting for niche adopters but not yet plug-and-play for broader audiences.
Prompt engineers, developers integrating Claude, AI researchers and power users who need stateful, governed AI workflows (Windows-focused)
Scatter-gather retrieval with prompt caching beats naive vector search for Claude Code memory.
First public implementation of DeepMind delegation paper. Tested on Zork with governance that blocked 'attack'.
Independent critic agent with no access to builder's reasoning catches blind spots.
Explainable retrieval with decision traces beats Mem0 and Zep on transparency.
Verifiable decision replay for autonomous systems, but execution complexity limits adoption beyond safety-critical domains.
Intercepts tool calls before execution to block dangerous actions like DB deletes.