This server provides a comprehensive, persistent, local-first project state management system backed by SQLite, with 82+ tools for AI coding assistants. It enables full lifecycle tracking and manipulation of tasks, decisions, artifacts, plans, observations, blockers, and milestones, along with their relationships (edges) featuring cycle detection. Key capabilities include:
Graph Operations: CRUD for nodes and edges, batch operations, and traversals.
Planning & Decomposition: Atomic feature planning with subtasks, milestones, and dependencies; auto-pruning of stale tasks; identification of next unblocked tasks.
Search & Querying: Full-text (FTS5) and semantic (TF-IDF) search, natural language queries, and raw SQL.
Time Travel & Versioning: Historical state snapshots, reverts, mutation history, undo, and diffing.
Spec-Driven Development: Ingest/export specs (Markdown, Gherkin), scaffold templates, verify acceptance criteria, and check compliance.
Multi-Agent Coordination: Ephemeral blackboard messaging for parallel subagents with channel/TTL support.
Analytics & Metrics: Velocity, burndown, cognitive load, ROI/synergy metrics, project summaries, and audits.
Sessions & Auditing: Session management, cryptographic event audit chains, post-mortems, and trajectory exports.
Git Integration: Branch sync, merge resolution, and branch-specific queries.
Import/Export: Graph export in JSON, DOT, Mermaid, HTML; issue sync with GitHub/Jira; database backup/restore/merge.
Maintenance & Health: Validation, auto-healing, compaction, pruning, doctor diagnostics, and cycle/orphan detection.
Automation & LLM Optimization: Context snapshots, change watching, and offloading dense state to the local graph to reduce token usage.
Visualization & Integration: Export graphs for 3D visualizers, link visual memory states, and scaffold FDD/RFC templates.
All operations are transactional with deterministic performance, suitable for single-player or multi-agent workflows.
Provides Git-aware state tracking by dynamically scoping project state to the current Git branch, and offers VCS branch sync and merge resolution tools for the state graph.
@putervision/state-memory-mcp is a zero-infrastructure, deterministic Model Context Protocol (MCP) server that provides AI coding assistants (such as Cursor, Claude Code, Gemini, or Copilot) with a structured, persistent SQLite graph for tracking workflow state—tasks, decisions, artifacts, plans, blockers, and their semantic relationships.
🌐 Official Documentation & Website: statememorymcp.com
Prerequisites: Node.js >= 20.0.0
# 1. Install globally
npm install -g @putervision/state-memory-mcp
# 2. Navigate to your project directory
cd your-project
# 3. Initialize state-memory-mcp
# Creates .state-memory-mcp/, updates .gitignore, registers project,
# and scaffolds IDE instructions and MCP configs for Cursor, Claude, VS Code, Windsurf, etc.
state-memory-mcp init
# Done! Restart your IDE or Agent Manager to activate.# Run directly via binary (after global install)
state-memory-mcp run
# Re-initialize across all registered workspace projects
state-memory-mcp init-globalRelated MCP server: contextforge-mcp
🧠 Deterministic State Memory: Zero LLM in the loop for memory operations; fast, deterministic SQLite graph traversals.
⚡ 82 Core MCP Tools: Full CRUD, relationship linking, DAG cycle checks, FTS5 search, TF-IDF RAG, time-travel history rollback, Spec-Driven Development, and auto-healing validation.
📉 Up to 462× Token Savings: Offloads context to a local SQLite database, avoiding context bloat and linear prompt degradation.
🚀 67%–74% Latency Reduction: Eliminates multi-step file scanning loops; agents retrieve unblocked tasks and blockers in milliseconds.
🤝 Multi-Agent Blackboard: Shared Context Store allowing parallel subagents to publish decisions, tasks, and blocker updates safely.
🎨 Interactive 3D Visualizer: Browser-based dark-mode 3D WebGL force-directed graph visualizer (state-memory-mcp view).
🛡️ 100% Local & Private: Local-first architecture; all state stays inside .state-memory-mcp/ in your workspace.
Explore dedicated guides and deep dives in the docs/ directory:
Guide | Description |
Cognitive Externalization, FSM Formalism, First-Hop Determinism & Benchmark metrics. | |
Node Types ( | |
Auto-Initialization details, Environment Variables table, and Editor Configs (Cursor, VS Code, Claude, Antigravity, Windsurf). | |
CLI flags ( | |
Session Lifecycle, Event Audit Trail, Snapshots, Trajectories, Sub-directory support & Spec-Driven Development. | |
Complete reference for all 82 MCP Tools, read-only | |
Formal parameters, return schemas, and code signatures for all MCP endpoints. | |
Viewing and exporting the interactive WebGL 3D Force-Directed Graph visualizer. | |
SQLite tables, columns, indexes, and schema migration history. |
# Run full unit, integration, and performance benchmark test suite
npm run testDeveloped and maintained by PuterVision LLC. Released under the MIT License.
Local Storage Guarantee: All graph data, decision records, and event logs remain 100% local in your workspace. No telemetry or project data is ever transmitted.
Trademarks & Non-Affiliation: Product names (Cursor, Claude Code, Gemini, Windsurf, VS Code, GitHub, SQLite) are property of their respective owners and used solely for compatibility identification.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
User-owned memory for AI agents, Copilot, Claude, IDEs, CLIs, and chat apps over remote MCP.
Persistent memory and cross-session learning for AI coding assistants (hosted remote MCP).
Private-by-default, local-first memory/context/task orchestrator for MCP apps and agents.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/putervision/state-memory-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server