codebase_search (WarpGrep exploration subagent), edit_file (10,500 tok/s partial edits), and Reflex read-back tools that pull labeled production traffic — zero code changes.
One command to install:
Logged in? Your API key auto-fills above. Otherwise, get it from your dashboard.
Installation (All Clients)
1
1. Get Your API Key
Get your API key from the dashboard.
2
2. Configure Your MCP Client
- Claude Code
- Codex
- Cursor
- Claude Desktop
- VS Code
- Manual
One-liner Installation (Recommended):Manual Config File Method:Create or edit
Configure Claude to prefer Morph: Add this to your global Claude config:
.claude.json in your workspace:3
3. Test Installation
Claude Code: Type
Codex: Run
Cursor/VS Code: Make any code edit request - should use Morph automatically
Manual: Check server logs show “MCP Server started successfully”
/mcp and /tools to see Morph’s edit_file toolCodex: Run
codex mcp list to verify server is configured, then make edit requestsCursor/VS Code: Make any code edit request - should use Morph automatically
Manual: Check server logs show “MCP Server started successfully”
Configuration
Advanced Configuration
Custom API endpoint — For enterprise deployments or custom authentication flows:
/v1/chat/completions with the token in the Authorization: Bearer header. Forward these to https://api.morphllm.com/v1/chat/completions after handling auth/billing.
Warp Grep timeout — Increase for large codebases or slow networks:
Available Tools
codebase_search, github_codebase_search, and the three Reflex tools (list_reflexes, reflex_summary, get_reflex_traces) are enabled out of the box. edit_file ships disabled to avoid conflicting with client-native editors — opt in by setting DISABLED_TOOLS="" (or any value that doesn’t include edit_file) in the MCP server env.
The Reflex tools read production results back: which classifiers run on your account, firing rates per traced model (canary vs. incumbent), and the individual conversations that fired. They need traced traffic to read — see the tracing guide.
Troubleshooting
Server won’t start: Check API key, Node.js 16+, runnpm cache clean --forceTools missing: Restart client, validate JSON config
Workspace issues: Add
.git or package.json, or set WORKSPACE_MODE="false"Slow performance: Use
edit_file over write_file, check network to api.morphllm.com
Performance Optimization
Best Practices
- Use
edit_filefor modifications: Much faster than reading + writing entire files - Minimize edit scope: Include only the sections that need changes
- Batch related edits: Make multiple changes in a single
edit_filecall