state-memory-mcp

# PuterVision — MCP Research & Development Labs

> PuterVision is an AI engineering research laboratory pioneering deterministic multi-modal state memory, visual perception caching, 3D spatial world models, high-frequency behavior tree execution engines, and BDI reasoning architectures for autonomous AI developer agents and the Model Context Protocol (MCP).

## Core Research Focus
- **Deterministic Context Stores**: SQLite-backed state graph memory for AI coding agents to eliminate LLM hallucination in task tracking and audit trails.
- **Perceptual Fast-Path Matching**: Sub-5ms screen state recognition using Hamming distance dHash algorithms to bypass expensive vision LLM calls.
- **3D Spatial Memory & Simulation**: Durable entity tracking, exponential confidence decay, and AABB collision avoidance for game and desktop automation.
- **High-Frequency (~60Hz) Behavior Tree Runtime**: Deterministic in-browser tick loops with reactive trigger preemption and safety guardrails.
- **Strategic BDI Reasoning**: Multi-attribute expected utility evaluation, hierarchical goal decomposition, and explainable decision traces.

## Published Open-Source Initiatives (The PuterVision MCP Pentad)

### 1. State Memory MCP (`@putervision/state-memory-mcp`)
- **Website**: https://statememorymcp.com
- **Repository**: https://github.com/putervision/state-memory-mcp
- **Description**: Zero-infrastructure, deterministic Model Context Protocol (MCP) server providing AI agents with persistent SQLite state graph memory for tracking tasks, decisions, blockers, plans, and artifacts.
- **Primary Capabilities**: SQLite graph database, 3D WebGL graph visualizer, Spec-Driven Development (SDD), Git branch isolation, SHA-256 Merkle event ledger.

### 2. Visual Memory MCP (`@putervision/vision-memory-mcp`)
- **Website**: https://visionmemorymcp.com
- **Repository**: https://github.com/putervision/vision-memory-mcp
- **Description**: Visual perception cache & perceptual hashing MCP server. Caches webpage and app screenshots locally using dHash, local CLIP vector search, and accessibility trees to cut LLM token costs by up to 90%.
- **Primary Capabilities**: L1/L2 dHash zero-token fast-path, L3 local CLIP vector search, visual layout drift verification, WebM/MP4 video frame timeline ingestion.

### 3. Spatial World Model MCP (`@putervision/world-model-mcp`)
- **Website**: https://worldmodelmcp.com
- **Repository**: https://github.com/putervision/world-model-mcp
- **Description**: Deterministic 3D/2D spatial internal world model for autonomous agents with entity permanence, movement simulation, and collision avoidance.
- **Primary Capabilities**: Entity CRUD, topological spatial graphs, frustum ray-AABB occlusion culling, waypoint navigation, Playwright 3D game automation inputs.

### 4. Behavior Runtime MCP (`@putervision/behavior-mcp`)
- **Website**: https://behaviormcp.com
- **Repository**: https://github.com/putervision/behavior-mcp
- **Description**: High-frequency (~60Hz) in-browser behavior tree tactical runtime engine with reactive trigger interrupts and multi-layer safety guardrails.
- **Primary Capabilities**: Fail-closed behavior trees, 60 actions/sec rate limiter, leaf policy gates, watchdog timers, deterministic frame recordings.

### 5. Agent Reasoning MCP (`@putervision/agent-reasoning-mcp`)
- **Website**: https://agentreasoningmcp.com
- **Repository**: https://github.com/putervision/agent-reasoning-mcp
- **Description**: Strategic BDI reasoning and decision intelligence engine for autonomous AI agents — goals, utility evaluation, and reactive replanning.
- **Primary Capabilities**: Goal DAG decomposition, multi-attribute expected utility theory, exponential belief decay ($C = C_0 e^{-\lambda t}$), intention dispatch queue, explainable decision traces.

### 6. ScreenChunk (`ScreenChunk`)
- **Website**: https://screenchunk.com
- **Description**: Ultra-fast live screen capture studio, 2D region cropping, and WebRTC screen recorder with client-side zero-knowledge AES-256-GCM chunk protection.

### 7. BassMusic.AI (`BassMusic.AI`)
- **Website**: https://bassmusic.ai
- **Repository**: https://github.com/putervision/bassmusicai
- **Description**: Free, 100% client-side AI electronic bass music and vocal stem generator running in-browser via WebGPU and WebAssembly SIMD neural models.

## Canonical URLs & References
- **PuterVision R&D Labs**: https://putervision.com
- **State Memory MCP**: https://statememorymcp.com
- **Visual Memory MCP**: https://visionmemorymcp.com
- **Spatial World Model MCP**: https://worldmodelmcp.com
- **Behavior Runtime MCP**: https://behaviormcp.com
- **Agent Reasoning MCP**: https://agentreasoningmcp.com
- **ScreenChunk**: https://screenchunk.com
- **BassMusic.AI**: https://bassmusic.ai
- **GitHub Organization**: https://github.com/putervision