superlocalmemory 3.4.16 → 3.4.18

package/docs/v2-archive/MCP-MANUAL-SETUP.md

@@ -1,775 +0,0 @@
-# SuperLocalMemory V2 - Manual MCP Setup Guide
-
-This guide shows how to manually add SuperLocalMemory V2 as an MCP server in various applications and tools.
-
-**Auto-configured by install.sh / npm install:**
-- ✅ Claude Desktop, Cursor, Windsurf, Continue.dev
-- ✅ OpenAI Codex CLI, VS Code/Copilot, Zed, OpenCode
-- ✅ Antigravity, Perplexity, Gemini CLI
-- ⚠️ ChatGPT (HTTP transport — run `slm serve`, see below)
-- ⚠️ JetBrains (manual GUI setup — see below)
-- ⚠️ Cody (manual setup — see below)
-
-For tools marked ⚠️, follow the manual setup instructions below.
-
----
-
-## Table of Contents
-
-1. [Prerequisites](#prerequisites)
-2. [Configuration Template](#configuration-template)
-3. [IDE/App Specific Setup](#ideapp-specific-setup)
-   - [Claude Desktop](#claude-desktop)
-   - [Cursor](#cursor)
-   - [Windsurf](#windsurf)
-   - [Continue.dev (VS Code)](#continuedev-vs-code)
-   - [OpenAI Codex CLI](#openai-codex-cli)
-   - [VS Code / GitHub Copilot](#vs-code--github-copilot)
-   - [Gemini CLI](#gemini-cli)
-   - [JetBrains IDEs](#jetbrains-ides)
-   - [Cody (VS Code/JetBrains)](#cody-vs-codejetbrains)
-   - [ChatGPT Desktop App](#chatgpt-desktop-app)
-   - [Perplexity](#perplexity)
-   - [Zed Editor](#zed-editor)
-   - [HTTP Transport (Remote Access)](#http-transport-remote-access)
-   - [Custom MCP Clients](#custom-mcp-clients)
-4. [Troubleshooting](#troubleshooting)
-5. [Verification](#verification)
-
----
-
-## Prerequisites
-
-1. **SuperLocalMemory V2 installed:**
-   ```bash
-   ./install.sh
-   ```
-
-2. **MCP Python package installed:**
-   ```bash
-   pip3 install mcp
-   ```
-
-3. **Know your Python path:**
-   ```bash
-   which python3
-   # Output: /usr/bin/python3 or /opt/homebrew/bin/python3 or similar
-   ```
-
-4. **Know your install directory:**
-   ```bash
-   echo ~/.superlocalmemory
-   # Default: /Users/yourusername/.claude-memory
-   ```
-
----
-
-## Configuration Template
-
-This is the base MCP configuration that works across all tools:
-
-```json
-{
-  "mcpServers": {
-    "superlocalmemory-v2": {
-      "command": "python3",
-      "args": [
-        "/Users/yourusername/.claude-memory/mcp_server.py"
-      ],
-      "cwd": "/Users/yourusername/.claude-memory",
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      },
-      "description": "SuperLocalMemory V2 - 100% local memory system"
-    }
-  }
-}
-```
-
-**Replace `/Users/yourusername/.claude-memory` with your actual install path.**
-
----
-
-## IDE/App Specific Setup
-
-### Claude Desktop
-
-**Configuration File:** `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS)
-**Configuration File:** `%APPDATA%\Claude\claude_desktop_config.json` (Windows)
-**Configuration File:** `~/.config/Claude/claude_desktop_config.json` (Linux)
-
-**Steps:**
-
-1. Open Claude Desktop settings or directly edit the config file
-2. Add the MCP server configuration:
-
-```json
-{
-  "mcpServers": {
-    "superlocalmemory-v2": {
-      "command": "python3",
-      "args": [
-        "/Users/yourusername/.claude-memory/mcp_server.py"
-      ],
-      "cwd": "/Users/yourusername/.claude-memory",
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      },
-      "description": "SuperLocalMemory V2 - Local memory system"
-    }
-  }
-}
-```
-
-3. **Restart Claude Desktop completely**
-4. Verify: Look for "superlocalmemory-v2" in available tools
-
-**Auto-configured by install.sh:** ✅ Yes
-
----
-
-### Cursor
-
-**Configuration File:** `~/.cursor/mcp_settings.json` (macOS/Linux)
-**Configuration File:** `%USERPROFILE%\.cursor\mcp_settings.json` (Windows)
-
-**Steps:**
-
-1. Create or edit `~/.cursor/mcp_settings.json`:
-
-```json
-{
-  "mcpServers": {
-    "superlocalmemory-v2": {
-      "command": "python3",
-      "args": [
-        "/Users/yourusername/.claude-memory/mcp_server.py"
-      ],
-      "cwd": "/Users/yourusername/.claude-memory",
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      },
-      "description": "SuperLocalMemory V2"
-    }
-  }
-}
-```
-
-2. **Completely quit Cursor (Cmd+Q on macOS)** and reopen
-3. Go to Settings → Tools & MCP
-4. Verify "superlocalmemory-v2" shows with green indicator
-
-**Auto-configured by install.sh:** ✅ Yes
-
----
-
-### Windsurf
-
-**Configuration File:** `~/.windsurf/mcp_settings.json` (macOS/Linux)
-**Configuration File:** `%USERPROFILE%\.windsurf\mcp_settings.json` (Windows)
-
-**Steps:**
-
-1. Create or edit `~/.windsurf/mcp_settings.json`:
-
-```json
-{
-  "mcpServers": {
-    "superlocalmemory-v2": {
-      "command": "python3",
-      "args": [
-        "/Users/yourusername/.claude-memory/mcp_server.py"
-      ],
-      "cwd": "/Users/yourusername/.claude-memory",
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      },
-      "transport": "stdio",
-      "description": "SuperLocalMemory V2"
-    }
-  }
-}
-```
-
-2. Restart Windsurf
-3. Check MCP servers in settings
-
-**Auto-configured by install.sh:** ✅ Yes
-
----
-
-### Continue.dev (VS Code)
-
-**Configuration File:** `~/.continue/config.yaml` (macOS/Linux)
-**Configuration File:** `%USERPROFILE%\.continue\config.yaml` (Windows)
-
-**Steps:**
-
-1. Open VS Code → Continue extension settings
-2. Or edit `~/.continue/config.yaml` directly:
-
-```yaml
-mcpServers:
-  - name: superlocalmemory-v2
-    command: python3
-    args:
-      - /Users/yourusername/.claude-memory/mcp_server.py
-    cwd: /Users/yourusername/.claude-memory
-    env:
-      PYTHONPATH: /Users/yourusername/.claude-memory
-    description: SuperLocalMemory V2 - Local memory system
-```
-
-3. Reload VS Code window (Cmd+Shift+P → Reload Window)
-4. Verify in Continue extension panel
-
-**Auto-configured by install.sh:** ✅ Partial (manual merge if config exists)
-
----
-
-### OpenAI Codex CLI
-
-**Configuration File:** `~/.codex/config.toml`
-**Format:** TOML (not JSON)
-
-**Method 1 — CLI command (preferred):**
-
-```bash
-codex mcp add superlocalmemory-v2 \
-  --env "PYTHONPATH=/Users/yourusername/.claude-memory" \
-  -- python3 /Users/yourusername/.claude-memory/mcp_server.py
-```
-
-**Method 2 — Manual TOML config:**
-
-Add to `~/.codex/config.toml`:
-
-```toml
-[mcp_servers.superlocalmemory-v2]
-command = "python3"
-args = ["/Users/yourusername/.claude-memory/mcp_server.py"]
-
-[mcp_servers.superlocalmemory-v2.env]
-PYTHONPATH = "/Users/yourusername/.claude-memory"
-```
-
-**Verify:** `codex mcp list` should show `superlocalmemory-v2`
-
-**Auto-configured by install.sh:** ✅ Yes
-
----
-
-### VS Code / GitHub Copilot
-
-**Configuration File:** `~/.vscode/mcp.json` (user-level) or `.vscode/mcp.json` (workspace-level)
-
-**Important:** VS Code uses `"servers"` (not `"mcpServers"`) and requires `"type": "stdio"`.
-
-**Steps:**
-
-1. Create or edit `~/.vscode/mcp.json`:
-
-```json
-{
-  "servers": {
-    "superlocalmemory-v2": {
-      "type": "stdio",
-      "command": "python3",
-      "args": ["/Users/yourusername/.claude-memory/mcp_server.py"],
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      }
-    }
-  }
-}
-```
-
-2. Restart VS Code
-3. Copilot Chat should now have access to SuperLocalMemory tools
-
-**Auto-configured by install.sh:** ✅ Yes
-
----
-
-### Gemini CLI
-
-**Configuration File:** `~/.gemini/settings.json`
-
-**Steps:**
-
-1. Create or edit `~/.gemini/settings.json`:
-
-```json
-{
-  "mcpServers": {
-    "superlocalmemory-v2": {
-      "command": "python3",
-      "args": ["/Users/yourusername/.claude-memory/mcp_server.py"],
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      }
-    }
-  }
-}
-```
-
-2. Restart Gemini CLI
-3. Memory tools will be available in conversations
-
-**Auto-configured by install.sh:** ✅ Yes
-
----
-
-### JetBrains IDEs
-
-**Supported:** IntelliJ IDEA, PyCharm, WebStorm, GoLand, and all JetBrains IDEs with AI Assistant (2025.2+)
-
-**Steps (GUI-based):**
-
-1. Open your JetBrains IDE
-2. Go to **Settings → AI Assistant → MCP Servers**
-3. Click **"+"** to add a new server
-4. Enter:
-   - **Name:** `superlocalmemory-v2`
-   - **Command:** `python3`
-   - **Arguments:** `/Users/yourusername/.claude-memory/mcp_server.py`
-   - **Working Directory:** `/Users/yourusername/.claude-memory`
-   - **Environment Variables:** `PYTHONPATH=/Users/yourusername/.claude-memory`
-5. Click **OK** and restart
-
-**JSON template** (for reference): See `configs/jetbrains-mcp.json`
-
-```json
-{
-  "mcpServers": {
-    "superlocalmemory-v2": {
-      "command": "python3",
-      "args": ["/Users/yourusername/.claude-memory/mcp_server.py"],
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      }
-    }
-  }
-}
-```
-
-**Auto-configured by install.sh:** ❌ No (manual GUI setup required)
-
----
-
-### Cody (VS Code/JetBrains)
-
-**Configuration:** Via VS Code settings or IDE settings
-
-**Steps:**
-
-1. Open VS Code → Settings → Extensions → Cody
-2. Look for "MCP Servers" or "External Tools" section
-3. Add server configuration:
-
-```json
-{
-  "cody.mcpServers": [
-    {
-      "name": "superlocalmemory-v2",
-      "command": "python3",
-      "args": ["/Users/yourusername/.claude-memory/mcp_server.py"],
-      "cwd": "/Users/yourusername/.claude-memory",
-      "env": {
-        "PYTHONPATH": "/Users/yourusername/.claude-memory"
-      }
-    }
-  ]
-}
-```
-
-4. Restart VS Code
-5. Check Cody status bar for connected MCP servers
-
-**Auto-configured by install.sh:** ❌ No (manual setup required)
-
----
-
-### ChatGPT Desktop App / ChatGPT Connectors
-
-**Important:** ChatGPT requires HTTP transport, not stdio. You need to run a local HTTP server and expose it via a tunnel. As of v2.3.5, SuperLocalMemory includes `search(query)` and `fetch(id)` MCP tools required by OpenAI's MCP spec for ChatGPT Connectors and Deep Research.
-
-**Requirements:**
-- ChatGPT Plus, Team, or Enterprise plan
-- **Developer Mode** must be enabled in ChatGPT settings
-- A tunnel tool: `cloudflared` (recommended, free) or `ngrok`
-- Reference: https://platform.openai.com/docs/mcp
-
-**Available Tools in ChatGPT:**
-
-| Tool | Purpose |
-|------|---------|
-| `search(query)` | Search memories (required by OpenAI MCP spec) |
-| `fetch(id)` | Fetch a specific memory by ID (required by OpenAI MCP spec) |
-| `remember(content, tags, project)` | Save a new memory |
-| `recall(query, limit)` | Search memories with full options |
-
-**Step-by-Step Setup:**
-
-1. **Start the MCP HTTP server:**
-   ```bash
-   slm serve --port 8417
-   # or with streamable-http transport (ChatGPT 2026+):
-   slm serve --port 8417 --transport streamable-http
-   # or using Python directly:
-   python3 ~/.superlocalmemory/mcp_server.py --transport http --port 8417
-   ```
-
-2. **Expose via cloudflared tunnel** (in another terminal):
-   ```bash
-   # Install cloudflared (if not installed)
-   # macOS: brew install cloudflared
-   # Linux: sudo apt install cloudflared
-
-   cloudflared tunnel --url http://localhost:8417
-   ```
-   Cloudflared will output a URL like `https://random-name.trycloudflare.com`
-
-   **Alternative — ngrok:**
-   ```bash
-   ngrok http 8417
-   ```
-
-3. **Copy the HTTPS URL** from cloudflared/ngrok output
-
-4. **Add to ChatGPT as a Connector:**
-   - Open ChatGPT (desktop or web)
-   - Go to **Settings → Connectors** (or **Settings → Apps & Connectors → Developer Mode**)
-   - Click **"Add Connector"**
-   - Paste the HTTPS URL with the `/sse/` suffix:
-     ```
-     https://random-name.trycloudflare.com/sse/
-     ```
-   - Name it: `SuperLocalMemory`
-   - Click **Save**
-
-5. **Verify in a new chat:**
-   - Start a new conversation in ChatGPT
-   - Look for the SuperLocalMemory connector in the tool selector
-   - Try: "Search my memories for authentication decisions"
-   - ChatGPT will call `search()` and return your local memories
-
-**Important Notes:**
-- The `/sse/` suffix on the URL is **required** by ChatGPT's MCP implementation
-- 100% local — your MCP server runs on YOUR machine. The tunnel just makes it reachable by ChatGPT. Your data is served on demand and never stored by OpenAI beyond the conversation.
-- The tunnel URL changes each time you restart cloudflared (unless you set up a named tunnel)
-- For persistent URLs, configure a cloudflared named tunnel: `cloudflared tunnel create slm`
-
-**Auto-configured by install.sh:** ❌ No (requires HTTP transport + tunnel)
-
----
-
-### Perplexity
-
-**Note:** Perplexity AI may support MCP integration via their developer settings.
-
-**Steps:**
-
-1. Open Perplexity app
-2. Go to **Settings → Integrations** or **Settings → Developer**
-3. Look for **"Custom MCP Servers"** or **"External Tools"**
-4. Add configuration:
-
-```json
-{
-  "name": "SuperLocalMemory V2",
-  "command": "python3",
-  "args": ["/Users/yourusername/.claude-memory/mcp_server.py"],
-  "cwd": "/Users/yourusername/.claude-memory",
-  "env": {
-    "PYTHONPATH": "/Users/yourusername/.claude-memory"
-  }
-}
-```
-
-5. Save and restart Perplexity
-6. Check if MCP tools appear in conversation context
-
-**Auto-configured by install.sh:** ❌ No (manual setup via app settings)
-
-**Note:** MCP support in Perplexity may vary by version. Check their documentation.
-
----
-
-### Zed Editor
-
-**Configuration File:** `~/.config/zed/settings.json`
-
-**Steps:**
-
-1. Open Zed → Settings (Cmd+,)
-2. Click "Open Settings File" (JSON editor)
-3. Add MCP server configuration:
-
-```json
-{
-  "assistant": {
-    "mcpServers": {
-      "superlocalmemory-v2": {
-        "command": "python3",
-        "args": ["/Users/yourusername/.claude-memory/mcp_server.py"],
-        "cwd": "/Users/yourusername/.claude-memory",
-        "env": {
-          "PYTHONPATH": "/Users/yourusername/.claude-memory"
-        }
-      }
-    }
-  }
-}
-```
-
-4. Save settings (Cmd+S)
-5. Restart Zed
-6. Check assistant panel for available MCP tools
-
-**Auto-configured by install.sh:** ❌ No (manual setup required)
-
----
-
-### HTTP Transport (Remote Access)
-
-For any MCP client that requires HTTP transport (ChatGPT, remote access, web-based tools):
-
-**Start the HTTP server:**
-
-```bash
-# Using CLI wrapper
-slm serve --port 8001
-
-# Using Python directly
-python3 ~/.superlocalmemory/mcp_server.py --transport http --port 8001
-```
-
-**For local access:**
-```
-http://localhost:8001
-```
-
-**For remote access (ChatGPT, etc.):**
-
-```bash
-# Option 1: ngrok (free tier available)
-ngrok http 8001
-
-# Option 2: Cloudflare Tunnel (free)
-cloudflared tunnel --url http://localhost:8001
-```
-
-Then use the HTTPS URL from the tunnel in your MCP client.
-
-**Security note:** The server binds to localhost only. The tunnel is the only way external services can reach it. Your data never leaves your machine — it's served on demand.
-
----
-
-### Custom MCP Clients
-
-If you're building your own MCP client or integrating with a custom tool:
-
-**Python Example:**
-
-```python
-from mcp import Client
-
-client = Client()
-client.add_server(
-    name="superlocalmemory-v2",
-    command="python3",
-    args=["/Users/yourusername/.claude-memory/mcp_server.py"],
-    cwd="/Users/yourusername/.claude-memory",
-    env={"PYTHONPATH": "/Users/yourusername/.claude-memory"}
-)
-
-# Use the MCP server
-response = client.call_tool("remember", {"content": "Test memory"})
-print(response)
-```
-
-**HTTP Transport (for remote access):**
-
-Start MCP server in HTTP mode:
-```bash
-python3 ~/.superlocalmemory/mcp_server.py --transport http --port 8001
-```
-
-Then connect via HTTP:
-```
-http://localhost:8001
-```
-
----
-
-## Troubleshooting
-
-### Issue 1: "MCP server not found"
-
-**Solution:**
-1. Verify file exists:
-   ```bash
-   ls -la ~/.superlocalmemory/mcp_server.py
-   ```
-2. Check permissions:
-   ```bash
-   chmod +x ~/.superlocalmemory/mcp_server.py
-   ```
-
-### Issue 2: "ModuleNotFoundError: No module named 'mcp'"
-
-**Solution:**
-```bash
-pip3 install mcp
-# Or if using conda:
-conda install -c conda-forge mcp
-```
-
-### Issue 3: "Python not found"
-
-**Solution:**
-1. Check Python installation:
-   ```bash
-   which python3
-   ```
-2. Use full Python path in config:
-   ```json
-   "command": "/usr/bin/python3"
-   ```
-
-### Issue 4: "Cannot import SuperLocalMemory modules"
-
-**Solution:**
-1. Verify `cwd` and `PYTHONPATH` are set in config
-2. Check modules exist:
-   ```bash
-   ls ~/.superlocalmemory/memory_store_v2.py
-   ```
-
-### Issue 5: IDE doesn't show MCP tools
-
-**Solution:**
-1. **Completely restart the IDE** (not just reload)
-2. Check IDE's MCP settings/logs
-3. Test MCP server manually:
-   ```bash
-   python3 ~/.superlocalmemory/mcp_server.py
-   # Should start without errors
-   ```
-
-### Issue 6: "Config file is overwritten"
-
-**Solution:**
-- install.sh creates backups: `mcp_settings.json.backup.YYYYMMDD-HHMMSS`
-- Restore if needed:
-  ```bash
-  cp ~/.cursor/mcp_settings.json.backup.* ~/.cursor/mcp_settings.json
-  ```
-
----
-
-## Verification
-
-After setup, verify MCP integration works:
-
-### 1. Check MCP Server Status
-
-```bash
-python3 ~/.superlocalmemory/mcp_server.py
-```
-
-**Expected output:**
-```
-============================================================
-SuperLocalMemory V2 - MCP Server
-Version: 2.7.0
-============================================================
-
-Transport: stdio
-Database: /Users/yourusername/.claude-memory/memory.db
-
-MCP Tools Available:
-  - remember(content, tags, project, importance)
-  - recall(query, limit, min_score)
-  - search(query)          [ChatGPT Connector]
-  - fetch(id)              [ChatGPT Connector]
-  - list_recent(limit)
-  - get_status()
-  - build_graph()
-  - switch_profile(name)
-  - backup_status()        [Auto-Backup]
-  - memory_used(...)       [v2.7 Learning]
-  - get_learned_patterns(...)  [v2.7 Learning]
-  - correct_pattern(...)   [v2.7 Learning]
-
-...
-```
-
-### 2. Test in Your IDE/App
-
-Try these natural language commands:
-
-**Save a memory:**
-```
-Remember that SuperLocalMemory V2 MCP integration works perfectly
-```
-
-**Search memories:**
-```
-What do we know about MCP integration?
-```
-
-**Check status:**
-```
-What's the status of my memory system?
-```
-
-### 3. Verify Tools are Visible
-
-In your IDE/app, check:
-- **Cursor:** Settings → Tools & MCP → Should show "superlocalmemory-v2" (green)
-- **Claude Desktop:** Tools menu → Should list SuperLocalMemory tools
-- **Continue.dev:** Extension panel → MCP servers section
-- **ChatGPT:** Tool selector in chat → Look for memory tools
-
----
-
-## Available MCP Tools
-
-Once configured, these 12 tools are available:
-
-| Tool | Purpose | Example Usage |
-|------|---------|---------------|
-| `remember()` | Save memories | "Remember we use React hooks" |
-| `recall()` | Search memories | "What did we decide about authentication?" |
-| `list_recent()` | Recent memories | "Show me recent discussions" |
-| `get_status()` | System health | "How many memories do I have?" |
-| `build_graph()` | Build knowledge graph | "Build the knowledge graph" |
-| `switch_profile()` | Change profile | "Switch to work profile" |
-| `search()` | Search memories (OpenAI MCP spec) | Used by ChatGPT Connectors and Deep Research |
-| `fetch()` | Fetch memory by ID (OpenAI MCP spec) | Used by ChatGPT Connectors and Deep Research |
-| `backup_status()` | Auto-backup status | "What's the backup status?" |
-| `memory_used()` | Feedback for learning (v2.7) | Implicit — called when a recalled memory is used |
-| `get_learned_patterns()` | Retrieve learned patterns (v2.7) | "What patterns have you learned about me?" |
-| `correct_pattern()` | Correct a learned pattern (v2.7) | "I actually prefer Vue, not React" |
-
-**Note:** `search()` and `fetch()` are required by OpenAI's MCP specification for ChatGPT Connectors. They are available in all transports but primarily used by ChatGPT. The 3 learning tools (`memory_used`, `get_learned_patterns`, `correct_pattern`) require v2.7's optional learning dependencies.
-
-Plus **2 MCP prompts** and **6 MCP resources** for advanced use.
-
----
-
-## Need Help?
-
-1. **Check logs:** Most IDEs have MCP server logs in their developer tools
-2. **Test manually:** Run `python3 ~/.superlocalmemory/mcp_server.py` to see errors
-3. **GitHub Issues:** https://github.com/qualixar/superlocalmemory/issues
-4. **Documentation:** https://github.com/qualixar/superlocalmemory
-
----
-
-**100% local. 100% private. Works everywhere.**
-
-SuperLocalMemory V2 - by Varun Pratap Bhardwaj
-Licensed under AGPL-3.0