MCP Integration

Your coding agent forgets your codebase every time you close the tab.

You spent 20 minutes explaining your project structure to Claude. Tomorrow, it won't remember any of it. The MCP integration fixes that. One setup, and your coding agent remembers your architecture decisions, coding conventions, bug fixes, and project context across every session.

Works with Claude Desktop, Cline, Cursor, and any MCP-compatible agent.

Setup Guide General Quickstart

mcp-setup.sh

# Install the Hebbrix MCP Server
cd mcp
./quick_setup.sh

# Add to your Claude Desktop config (claude_desktop_config.json)
{
  "mcpServers": {
    "hebbrix": {
      "command": "python",
      "args": ["/path/to/hebbrix/mcp/server.py"],
      "env": {
        "HEBBRIX_API_KEY": "your_api_key"
      }
    }
  }
}

# Restart Claude Desktop. Done.
# Your coding agent now has persistent memory.

What your coding agent can do now

Four MCP tools, each one gives your agent a new superpower it didn't have before.

hebbrix_remember

Remember anything from the conversation

"Remember that we use PostgreSQL with Prisma ORM, and all API routes go through the middleware layer."

Your agent stores this context. Tomorrow, next week, next month. It's there.

hebbrix_search

Search across everything it has learned

"What database pattern do we use?"

5-layer search finds the relevant memories even if you phrase it differently than when you stored it.

hebbrix_recall

Get context for the current task

Automatically surfaces relevant project context when you start working on a file or feature.

No prompting needed. The agent knows what context to pull based on what you're doing.

hebbrix_forget

Remove outdated information

"Forget that we use REST. We migrated to GraphQL last sprint."

Clean up stale context so your agent doesn't give outdated advice.

A week with memory

What it actually feels like when your coding agent remembers.

Monday

You explain your project's auth flow to Claude. It stores the pattern: JWT tokens, refresh rotation, middleware validation.

Wednesday

You ask Claude to add a new protected endpoint. Without you explaining anything, it uses your JWT pattern, adds the middleware, follows your naming conventions.

Friday

A bug in the auth flow. Claude remembers the implementation from Monday, knows exactly where to look. Suggests a fix that matches your error handling style.

Next Monday

New sprint, same agent. It remembers the bug fix, the auth pattern, your coding style. Picks up right where you left off. No re-explaining needed.

Give your coding agent a memory

One config file. Five minutes. Your AI assistant will never forget your project again.

Setup Guide Get Your API Key