# projectmem MCP server

> Keep local memory of issues, attempts, fixes and decisions per project, and warn coding agents before they repeat a failed approach.

- Listing: https://mcp.tc/i/projectmem
- Connect: this is a local (stdio) server; install it on your machine (see Install). The listing link is a page, not an MCP endpoint.
- Type: local (stdio)
- Auth: no API key needed
- Category: [Developer Tools](https://mcp.tc/c/developer-tools)
- Vendor: riponcm
- Homepage: <https://projectmem.dev>
- Docs: <https://projectmem.dev/guide>
- Repository: <https://github.com/riponcm/projectmem>
- Package: pypi `projectmem`

## About

projectmem is a local-first memory layer for AI coding agents. It records typed events in a .projectmem/ folder inside each repository: issues, attempts, fixes, decisions and notes. Agents read this memory at the start of a session and log their work, and a pre-commit warning fires before an approach that already failed is repeated.

It runs as a local stdio MCP server installed from PyPI (pip install projectmem, or uvx). One server configuration serves every registered project. It works with Claude Code, Claude Desktop, Cursor, Antigravity and Codex. No account, API key or cloud service is needed, and there is no telemetry.

## What it can do

- Read a token-budgeted memory summary of a project, with optional focus filter
- Log issues, attempts, fixes and decisions as typed events
- Warn before repeating an approach that already failed
- Serve every registered project from a single MCP configuration
- Keep memory in a plain .projectmem/ directory inside the repo

## Example prompts

- "Load the projectmem context for this project before we start."
- "Log this failed attempt and the fix we found in projectmem."
- "Check projectmem for earlier attempts at this bug before I try a fix."
- "Record our decision to switch the cache layer as a projectmem decision."

## Install

### Claude Code

1. Run this in a terminal, in your project folder:

```bash
claude mcp add --transport stdio projectmem -- uvx projectmem
```

2. Start Claude Code and type `/mcp`. **projectmem** should show as connected.

Add `--scope user` to make it available in every project.

### Claude Desktop

1. Open **Settings → Developer → Edit Config**. It opens `claude_desktop_config.json`. Add:

`claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "projectmem": {
      "command": "uvx",
      "args": [
        "projectmem"
      ]
    }
  }
}
```

2. Save the file and restart Claude Desktop.

Needs uv (Python) on your computer. The file lives in `~/Library/Application Support/Claude/` on macOS and `%APPDATA%\Claude\` on Windows.

### Cursor

[Add to Cursor](<https://cursor.com/install-mcp?name=projectmem&config=eyJjb21tYW5kIjoidXZ4IiwiYXJncyI6WyJwcm9qZWN0bWVtIl19>) (opens Cursor)

Or add it by hand to `~/.cursor/mcp.json` (all projects) or `.cursor/mcp.json` (this project):

`mcp.json`:

```json
{
  "mcpServers": {
    "projectmem": {
      "command": "uvx",
      "args": [
        "projectmem"
      ]
    }
  }
}
```

Needs uv (Python) on your computer.

### VS Code

[Install in VS Code](<https://vscode.dev/redirect/mcp/install?name=projectmem&config=%7B%22type%22%3A%22stdio%22%2C%22command%22%3A%22uvx%22%2C%22args%22%3A%5B%22projectmem%22%5D%7D>) (opens VS Code)

Or from a terminal:

```bash
code --add-mcp '{"name":"projectmem","type":"stdio","command":"uvx","args":["projectmem"]}'
```

Or commit it to the repo in `.vscode/mcp.json`:

`.vscode/mcp.json`:

```json
{
  "servers": {
    "projectmem": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "projectmem"
      ]
    }
  }
}
```

Needs uv (Python) on your computer.

### Devin Desktop

1. Add it to `~/.config/devin/mcp_config.json` (macOS and Linux) or `%APPDATA%\devin\mcp_config.json` (Windows):

`mcp_config.json`:

```json
{
  "mcpServers": {
    "projectmem": {
      "command": "uvx",
      "args": [
        "projectmem"
      ]
    }
  }
}
```

2. Refresh the MCP server list in Cascade.

Devin Desktop is the new name for Windsurf.

### Codex

```bash
codex mcp add projectmem -- uvx projectmem
```

Or edit `~/.codex/config.toml` directly:

`config.toml`:

```toml
[mcp_servers.projectmem]
command = "uvx"
args = ["projectmem"]
```

Needs uv (Python) on your computer.

### Gemini CLI

```bash
gemini mcp add projectmem uvx projectmem
```

This adds it to the current project. Add `-s user` to use it everywhere.

### Any client

Most clients that start local servers accept this shape:

```json
{
  "mcpServers": {
    "projectmem": {
      "command": "uvx",
      "args": [
        "projectmem"
      ]
    }
  }
}
```

Zed puts servers under `context_servers` in its settings, with the same `command`, `args` and `env` fields.

Needs uv (Python) on your computer.

## Recent changes

- 2026-10-31 02:27 UTC: New version 0.3.5 (was 0.3.4), read from PyPI. Source: automatic check.
- 2026-10-11 04:27 UTC: Tracking started: version 0.3.4. Source: automatic check.

More on the History tab, with descriptions before and after for the newest entries: https://mcp.tc/i/projectmem#history (Atom feed: https://mcp.tc/i/projectmem/history.xml). It records what mcp.tc's anonymous checks saw, every 3 hours where the tool list can be read without sign-in; it is not a security review.

## Details

- Server version: 0.3.5
- Last checked: 2026-10-11
- Listed: 2026-10-11
- Updated: 2026-10-11

---
Source: https://mcp.tc/i/projectmem (mcp.tc is an independent directory, not affiliated with this server's publisher). Corrections: https://mcp.tc/report
