Model Context Protocol (MCP) tool to interact with Claude Desktop on macOS
Claude desktop mcp mcp server connects Claude Desktop MCP to AI assistants that speak the Model Context Protocol. Model Context Protocol (MCP) tool to interact with Claude Desktop on macOS.
A Model Context Protocol (MCP) server that enables Claude Code to communicate with Claude Desktop. This server allows Claude Code to send prompts to Claude Desktop and poll for responses.
Once connected, the assistant can call these 11 tools directly:
Tool — askParameters — { "prompt": "What is dependency injection?" }description — What changes to review (e.g., "authentication fix")polling_interval — How often to check for response (default: 1.5s)timeout — Maximum wait time for response (default: 30s)get_conversations — The get_conversations tool exposed by this serverWorkarounds — 1. Use Claude's API: For programmatic access to responses, consider using Claude's API directly instead of desktop automation 2. Manual verificationSetup — 1. Copy the example command to your Claude Commands directory: bash cp examples/claude-peer-review.md ~/.claude/commands/Usage — The peer review command accepts up to 3 arguments: - description: What changes to review (e.g., "authentication fix") - polling_interval: How often toLinting — The Linting tool exposed by this serverTools — The Tools tool exposed by this serverSetup follows the standard MCP pattern: clone or install the server, then register it in your client's configuration file and restart the client. The configuration snippets on this page cover Claude Desktop, Claude Code and Cursor.
The server reads 2 environment variables: SKIP_CLAUDE_POLLING, LOG_LEVEL. Keep credentials in your client's env block or a secrets manager rather than committing them.
File and storage servers are what separate an assistant that talks about your documents from one that actually works with them. Claude Desktop MCP sits in that group, and the shape of its toolset — Tool, Parameters, description among others — tells you what it is really for. Worth comparing against the other file systems servers in this directory before you commit to one, since several overlap in scope but differ sharply in setup cost and permissions.
| Tool | What it does |
|---|---|
| Tool | ask |
| Parameters | { "prompt": "What is dependency injection?" } |
| description | What changes to review (e.g., "authentication fix") |
| polling_interval | How often to check for response (default: 1.5s) |
| timeout | Maximum wait time for response (default: 30s) |
| get_conversations | The get_conversations tool exposed by this server. |
| Workarounds | 1. **Use Claude's API**: For programmatic access to responses, consider using Claude's API directly instead of desktop automation 2. **Manual verification**: After sending a prompt, manually check the Claude Desktop wind |
| Setup | 1. Copy the example command to your Claude Commands directory: bash cp examples/claude-peer-review.md ~/.claude/commands/ |
| Usage | The peer review command accepts up to 3 arguments: - **description**: What changes to review (e.g., "authentication fix") - **polling_interval**: How often to check for response (default: 1.5s) - **timeout**: Maximum wai |
| Linting | The Linting tool exposed by this server. |
| Tools | The Tools tool exposed by this server. |
{
"mcpServers": {
"claude-desktop": {
"command": "npx",
"args": ["mcp-claude-desktop"]
}
}
}Configuration as documented by the project. Restart the client after saving.
| Variable | Description | Required |
|---|---|---|
| SKIP_CLAUDE_POLLING | Configuration value read at startup. | Optional |
| LOG_LEVEL | Configuration value read at startup. | Optional |
A knowledge graph your assistant keeps between sessions — entities, relations and observations that persist.
Kill hallucinated APIs — version-accurate, up-to-date library documentation injected straight into context.
Your workspace, on speaking terms with AI — search, read and write Notion pages and databases.
A structured scratchpad for hard problems — stepwise reasoning with revisions, branches and visible logic.
Symbol-level code navigation, refactoring and memory for coding agents — the IDE brain your assistant has been missing.
Chat with your second brain — search, read and write vault notes through the Local REST API.