⚡ 40–70× Token Reduction

~800 tokens per code question instead of 50,000+. Real-world bills drop 40–70%.

🔒 100% Local & Offline

Your code never leaves your machine. Zero cloud APIs, zero telemetry.

✅ NIST AI RMF Compliant

Full audit trail and enterprise compliance reporting.

🛠️ Works Everywhere

Claude Code, Cursor, ChatGPT, Gemini, or any LLM.

How It Works

Phase 1: Smart Retrieval

A 4-layer semantic index surfaces only the context your question needs—not the entire codebase.

Phase 2: Output Compression

PostToolUse hooks compress Read/Bash/Grep output 88–91% smaller before agents see it.

Phase 3: Brain-like Memory (v0.4.0)

A persistent weighted graph runs alongside the LLM, learning Hebbian associations between co-active code nodes. Decay prunes stale links; long-term potentiation protects frequently-used ones; spreading activation surfaces related code on every prompt — no MCP call required.

Phase 4: Graph View (new)

neuralmind serve opens an Obsidian-style force-directed graph of your codebase in the browser. Structural edges, the learned synapse overlay, backlinks, a semantic quick-switcher, and one-click "open in editor". Makes the brain inspectable — no more black-box retrieval.

Result: 40–70× Reduction

50K+ tokens of raw source compressed to ~800 tokens of structured context per code question. Cost bills drop 40–70%. The brain-like layer makes retrieval get sharper the longer NeuralMind runs on a codebase.

Enterprise Features

NIST AI RMF Audit Trail

Every query logged. Export for compliance and auditor review.

Pluggable Backends

ChromaDB, PostgreSQL pgvector, or LanceDB. Choose your infrastructure.

MCP Security

RBAC, rate limiting, secret detection, audit logging.

SOC 2 & GDPR Ready

Full transparency, no vendor lock-in, zero exfiltration.

How NeuralMind Compares

Feature	NeuralMind	Cursor @codebase	Claude Projects	Long Context
Works everywhere	✅ Yes	❌ Cursor only	⚠️ Claude only	✅ Yes
100% local & offline	✅ Yes	❌ No	❌ No	❌ No
Token reduction	40–70×	2–3×	0× (loads all)	0× (loads all)
Enterprise compliance	✅ NIST AI RMF	⚠️ Basic	⚠️ Basic	⚠️ Basic

View detailed comparisons →

5-Minute Setup

# 1. Install NeuralMind
pip install neuralmind

# 2. Install graphify (required)
git clone https://github.com/safishamsi/graphify.git
cd graphify && pip install -e .

# 3. Go to your project
cd /path/to/your-project

# 4. Generate the code graph
graphify update .

# 5. Build the neural index
neuralmind build .

# 6. Test it
neuralmind stats .

Full Setup Guide

Use Cases

💰 Cost Optimization

Reduce monthly AI bills by 40–70% while improving answer quality.

🔐 Regulated Industries

Process sensitive code without exfiltration. GDPR/HIPAA compliant.

📈 Growing Monorepos

Scale AI help to 100K+ LOC without loading entire files.

⚡ Any LLM

ChatGPT, Gemini, local models, Claude Code. Works with all of them.

See detailed use cases →

Documentation & Resources

Setup Guide

First-time setup for all platforms.

Scheduling Guide

Auto-discovery, cloud sync, CI/CD.

CLI Reference

All commands and flags.

Enterprise Features

NIST AI RMF, audit trails, security.

Version Strategy

Versioning, support timeline, upgrades.

Troubleshooting

Common issues & solutions.

Browse All Docs

Status & Future

✅ v0.19.0 (Current) — MCP distribution

neuralmind install-mcp --all auto-detects installed agents (Claude Code, Cursor, Cline, Claude Desktop) and registers NeuralMind's MCP server with each — non-destructive, idempotent. The agent onboards through NeuralMind's tools instead of grepping cold.

📋 Next — measured onboarding-lift

The self-benchmark already measures the learned-synapse uplift (Phase 3: top-k hit rate 71.7% → 83.3% with recall on). Next: formalise the E1.5 onboarding-lift eval — a cold agent plus a committed team baseline vs a cold agent alone — as the headline differentiator.

🎯 v0.19 → v1.0

The moat (MCP distribution + a measured learned-synapse onboarding-lift), then a stable API guarantee and LTS toward v1.0.

See the next-release plan → · Roadmap →

🧠 NeuralMind

The Problem & Solution