@shivaclaw

privacy-preserving context stripper that removes soul, user, and memory metadata before cloud api calls, routing simple queries to local llms first and sanitizing cloud fallbacks. targets teams with existing local inference.

nirvana-plugin routes openclaw queries between local qwen2.5:7b inference and cloud apis, enforcing privacy via context stripping and audit logging. bundled model requires docker/ollama, aims for 80%+ local handling with seamless fallback.

Local-first privacy-first inference. Your OpenClaw agent thinks locally and asks the cloud intelligently. Saves 85%+ tokens, protects privacy, agent learns f...

Permanent memory for OpenClaw agents. Lossless capture, intelligent routing, semantic recall, and disaster recovery in five tiers.

Give your AI agent genuine continuity, identity, and resilience across every session and deployment with a four-tier persistent memory stack.