MemVault MCP Server - Connect Claude Desktop to your code memory SaaS
npm install memvault-mcp!Build Status
!License
!NPM Version
!Marketplace
> Give your LLMs long-term memory, semantic understanding, and evolving context—with one line of code.
MemVault is a production-grade GraphRAG (Graph Retrieval-Augmented Generation) platform. Unlike simple vector databases that only find "similar words", MemVault builds a dynamic knowledge graph of entities and relationships, allowing your AI to understand context, not just keywords.
Start 7-Day Free Trial | Read Documentation | NPM SDK
---
Building persistent memory is hard. Managing vector databases, embedding pipelines, graph databases, and context windows is even harder. MemVault solves this with a managed API that acts as the hippocampus for your AI agents.
* Ingest Now, Process Later: We accept data instantly, but deep processing happens in the background.
* Auto-Consolidation: Every 6 hours, our Sleep Cycle Engine wakes up to merge duplicate entities, prune noise, and strengthen semantic relationships in the graph.
* Result: Your AI gets smarter over time without you writing a single line of maintenance code.
pgvector semantic search with keyword extraction for maximum retrieval accuracy.---
``bash`
npm install memvault-sdk-jakops88
`typescript
import { MemVault } from 'memvault-sdk-jakops88';
// Initialize with your 'sk_...' key from the dashboard
const memory = new MemVault({
apiKey: process.env.MEMVAULT_API_KEY
});
// 1. Store a memory (Text -> Vector + Graph Node)
await memory.add({
content: "The user, Jakob, is a Senior Developer who prefers TypeScript over Python.",
tags: ["user-profile", "preferences"]
});
// 2. Ask questions (Retrieves context via GraphRAG)
const context = await memory.search("What is Jakob's preferred language?", {
limit: 1,
strategy: 'hybrid' // Uses both Vector and Graph traversal
});
console.log(context);
// Output: "Jakob prefers TypeScript (Confidence: 98%)"
`
---
Keep your AI updated with your codebase automatically. Use our official Action to sync documentation or code files directly to your MemVault knowledge graph on every push.
`yaml.github/workflows/memvault-sync.yml
name: Sync Docs to Brain
on:
push:
paths: ['docs/**']
jobs:
sync:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: MemVault Sync
uses: jakops88-hub/long-term-memory-api/.github/actions/memvault-sync@v1
with:
memvault_api_key: ${{ secrets.MEMVAULT_API_KEY }}
file_paths: "docs/*/.md"
`
---
MemVault is built for scale and security, hosted on high-performance infrastructure.
* API Layer: Node.js/Express with strict rate limiting and validation (Zod).
* Async Workers: Heavy lifting (Graph extraction, Sleep Cycles) is offloaded to Redis/BullMQ queues to ensure sub-millisecond API response times.
* Storage: PostgreSQL with pgvector for high-dimensional vector storage.
* Security: All keys are encrypted. Usage is sandboxed per user via HybridAuth.
---
We offer a straightforward pricing model designed for developers.
| Plan | Price | Features |
|------|-------|----------|
| Trial | Free (7 Days) | Full access to Hobby tier to test the API. |
| Hobby | $29/mo | 100k tokens, GraphRAG, Dashboard access. Hard limits (no overage). |
| Pro | $99/mo | 1M tokens, Sleep Cycles (Consolidation), Priority Support. |
---
MemVault is Open Core. You can run the stack locally for development or compliant internal usage. Note that Sleep Cycles and Cost Guard are optimized for the managed cloud environment.
Click to view Docker instructions
bash
git clone https://github.com/jakops88-hub/long-term-memory-api.git
`
2. Configure .env:
`bash
cp .env.example .env
# Add your DATABASE_URL and OPENAI_API_KEY
`
3. Run with Docker Compose:
`bash
docker-compose up -d
`
The API is now available at http://localhost:3000`.---
MIT © Jakob Sandström