Skip to content
MCP server

jevmem

Project rules kept in a JEVMEM.md file in git, with a guard on risky commands

Runs locallyStatus and history
Pricing
Open source
Platforms
CLI, macOS, Windows
Runs
Locally
Version
0.7.8
Listing updated
3 days ago

About jevmem

jevmem is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Decisions, rules and failed approaches from your chats are saved to a single JEVMEM.md in the repository. Teammates receive the same file through git, and Claude Code, Cursor and Codex all read it. It sits beside Claude Code's own auto memory and adds a shared file, support for other tools and the guard.

The guard works before an action: if a rule says generated migration files must not be hand-edited, Claude Code stops and asks you, quoting the rule, before an edit to one of them runs. The MCP server adds add_memory, search_memory and audit_memory, which call TypeSafe AI's Jev and need a TypeSafe API key.

It runs locally over stdio and one server entry serves one project. It is MIT licensed and listed in Anthropic's Claude plugin directory.

Install

Run in a terminal

claude mcp add jevmem --env TYPESAFE_API_KEY='<TYPESAFE_API_KEY>' -- npx -y jevmem

Add to claude_desktop_config.json

{
  "mcpServers": {
    "jevmem": {
      "command": "npx",
      "args": [
        "-y",
        "jevmem"
      ],
      "env": {
        "TYPESAFE_API_KEY": "<TYPESAFE_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "jevmem": {
      "command": "npx",
      "args": [
        "-y",
        "jevmem"
      ],
      "env": {
        "TYPESAFE_API_KEY": "<TYPESAFE_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "jevmem": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "jevmem"
      ],
      "env": {
        "TYPESAFE_API_KEY": "<TYPESAFE_API_KEY>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "jevmem": {
      "command": "npx",
      "args": [
        "-y",
        "jevmem"
      ],
      "env": {
        "TYPESAFE_API_KEY": "<TYPESAFE_API_KEY>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "jevmem": {
      "command": "npx",
      "args": [
        "-y",
        "jevmem"
      ],
      "env": {
        "TYPESAFE_API_KEY": "<TYPESAFE_API_KEY>"
      }
    }
  }
}

Configuration

  • TYPESAFE_API_KEYTypeSafe API key for Jev, which add_memory, search_memory and audit_memory use. If unset, the server reads it from <project>/.jevmem/.env or ~/.jevmem/env.
  • JEVMEM_ROOTProject root, for clients that start the server without a project working directory (Claude Desktop). One server entry serves one project.
  • TYPESAFE_BASE_URLSend Jev requests through a proxy or gateway instead of api.typesafe.ai.
  • JEVMEM_CACHE0 turns off the answer cache in .jevmem/.

Tools this server provides (3)

Taken from the project documentation; the running server may differ.

add_memory
Saves a decision, rule or failed approach to project memory.
search_memory
Searches saved project memory.
audit_memory
Audits saved memory against the project.

What it does

Rules file in git
One JEVMEM.md carries rules and decisions to every teammate.
Command guard
Checks Claude's actions against saved rules and asks before breaking one.
Cross-tool reading
Claude Code, Cursor and Codex read the same file.
Memory search and audit
Tools for saving, searching and auditing project memory.
Answer cache
A local cache in .jevmem/ can be turned off with JEVMEM_CACHE=0.

Good for

  • Stop agents from editing generated files
  • Share team decisions across several coding agents
  • Record failed approaches so they are not retried
  • Audit saved project rules for conflicts

Pricing

Open source

MIT licensed; memory tools use a TypeSafe AI API key

Prices change. Check avinash-jetwani.github.io before you buy.

jevmem FAQ

Is it open source?

Yes, MIT licensed.

What does the memory search need?

A TypeSafe API key, set as TYPESAFE_API_KEY or read from .jevmem/.env or ~/.jevmem/env.

Does it work in Claude Desktop?

Yes over MCP; set JEVMEM_ROOT since Claude Desktop starts servers without a project directory.

Can one server cover many projects?

No, one server entry serves one project.

Is jevmem free?

Yes. jevmem is open source, so you can use and self-host it for free.

How do I add jevmem to Claude Code?

Run this command in your terminal: claude mcp add jevmem --env TYPESAFE_API_KEY='<TYPESAFE_API_KEY>' -- npx -y jevmem

What are the best alternatives to jevmem?

The closest alternatives to jevmem are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to jevmem

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools