How MemNexus Works
Sign up, install the CLI, and run one command. Your coding agents get persistent memory that carries context across sessions, projects, and tools.
Sign Up and Install
Create an account at memnexus.ai and grab your API key from the customer portal. Then install the CLI:
npm install -g @memnexus-ai/cliRun Setup
mx setup is an interactive CLI that detects your AI coding tools and configures everything. It walks you through:
mx setup- MCP server connection — writes the MemNexus server entry to your tool's config so it can communicate with your memory store
- Steering rules — adds rules that teach your agent when to search for context, when to save decisions, and how to format memories
- Slash commands — installs /save, /search, and /recall for direct control over memory operations
- CommitContext — sets up a git hook that captures the reasoning behind every commit into your memory store
Create Your First Memory
In your coding agent, tell it something worth remembering. MemNexus stores it with full semantic indexing — topics, entities, and facts are automatically extracted.
"Save a memory: we chose Postgres over MongoDB because our access patterns are relational and the ops team already runs Postgres."
Search in a New Session
Close the conversation. Open a new one. Ask your agent to search. It queries your memory store via MCP and surfaces the decision with its full reasoning — even in a completely new session.
"Search my memories for why we chose our database."
Browse in the Portal
Your memories are also browsable in the MemNexus customer portal. Search, filter by topic, view the knowledge graph that connects your decisions and entities, and see how your project knowledge grows over time.
Supported Coding Agents
One memory store across every tool you use. mx setup detects your installed agents and configures them automatically.
Claude Code
Persistent memory across sessions via MCP. Steering rules in CLAUDE.md. CommitContext captures reasoning on every commit.
Setup guide →Cursor
MCP server connection configured automatically. Steering rules in .cursor/rules/. Memory persists across Composer and Agent sessions.
Setup guide →GitHub Copilot
MCP integration via VS Code. Steering rules in .github/copilot-instructions.md. Works with Copilot Chat and the coding agent.
Setup guide →Windsurf
MCP server for Cascade. Steering rules in .windsurf/rules/. Memory carries across Cascade conversations.
Setup guide →Continue.dev
Native MCP support with stdio, SSE, and streamable HTTP. Steering rules in .continue/rules/. MCP tools available in Agent mode.
Setup guide →Kiro
Native MCP client. Steering files for project context. Memory complements Kiro's spec-driven development workflow.
Setup guide →JetBrains AI
MCP integration for IntelliJ, PyCharm, WebStorm, and all JetBrains IDEs. Steering via .jetbrains/ai.md prompts.
Setup guide →Cody
MCP support via Sourcegraph's extension. Memory adds temporal context on top of Cody's deep codebase intelligence.
Setup guide →Tabnine
MCP configured via .tabnine/mcp_servers.json. Memory layer for privacy-first teams using Tabnine Enterprise.
Setup guide →Cline
MCP integration in VS Code. Memory persists across Cline's autonomous coding sessions.
Setup guide →Aider
MCP integration for the terminal-based AI pair programmer. Memory carries context across aider sessions.
Setup guide →Zed
MCP support in Zed's built-in AI assistant. Memory adds persistence to Zed's fast, native editor experience.
Setup guide →MemNexus works with any MCP-compatible tool. Building a custom agent? Use the TypeScript SDK or REST API for direct integration.
Get started in under 5 minutes
Sign up, install the CLI, and run mx setup. Works with Claude Code, Cursor, GitHub Copilot, Windsurf, and more.
Already have an account? Sign in →