Read I/O locking using LRU cache
Retrieves & observes Kubernetes Pod resource (CPU, memory) utilisation.
OpenClaw extension that connects FlowState agents to the shared AgentMemory service
CLI to configure Maess Memory hooks and project metadata.
Claude Code forgets everything between sessions. deja fixes that.
Self-Optimizing Neural Architecture (SONA) - Runtime-adaptive learning with LoRA, EWC++, and ReasoningBank for LLM routers and AI systems. Sub-millisecond learning overhead, WASM and Node.js support.
moon command line and core system.
A powerful TypeScript framework for building event-driven, domain-driven applications with Event Sourcing and CQRS patterns
Add living Obsidian docs + Claude Code skill to any repo
Musubi memory plane for OpenClaw agents — episodic capture, curated knowledge, and presence-to-presence thoughts with authority-weighted prompt supplements.
MCP server for devsh/cmux agent memory - enables Claude Desktop and external clients to access sandbox memory and orchestrate multi-agent workflows
All-in-one personal AI agent — OpenClaw spine + Hermes brain + Cowork outcome mode, with ChatGPT-subscription login, a local OpenAI-compatible endpoint, and a clean web UI.
Analyize recursive (cyclic) JSON Schema types
LaunchDarkly Server-Side SDK for Vercel Edge
Async Redis client mock object for unit testing
A local DynamoDB server for testing and development. Provides both a fast in-memory implementation and a Java-based implementation using the official AWS DynamoDB Local.
OpenClaw-style memory system for OpenCode with full automation and real vector search
Hindsight memory plugin for OpenClaw - biomimetic long-term memory with fact extraction
Declarative control plane for agent harnesses — one YAML for grounding, tools, memory, and hooks.
Fast point-in-polygon intersection for large numbers of polygons.
Studio UI and HTTP runtime for Anvia agents.
Structured memory system for AI agents — typed storage, knowledge graph, context profiles, canvas dashboards, neural graph themes, and Obsidian-native task views. An elephant never forgets. 🐘
Intelligent LLM API proxy — prompt caching, smart routing, memory. Drop-in replacement that cuts inference costs 70-95%.
Secure, isolated expression evaluation runtime for n8n