What this error means
Claude Code MCP servers from ~/.claude/settings.json missing in desktop app is a Claude Code failure pattern reported for developers trying to fix claude code desktop app not loading mcp servers from settings.json. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.
Why this happens
main.log shows replaceRemoteMcpServers only loads 7 app-level connectors. settings.json MCP servers (ado, sql-server, figma) completely absent. CLI sessions load all MCP servers correctly. Desktop app v2.1.138 on Windows.
Common causes
- MCP servers configured in ~/.claude/settings.json are not started or made available in Claude Code desktop app local sessions. Only built-in app-level connectors load. The same config works perfectly in CLI sessions. Users cannot use their configured tools in the desktop app.
- main.log shows replaceRemoteMcpServers only loads 7 app-level connectors. settings.json MCP servers (ado, sql-server, figma) completely absent. CLI sessions load all MCP servers correctly. Desktop app v2.1.138 on Windows.
Quick fixes
- Confirm the exact error signature matches
Claude Code MCP servers from ~/.claude/settings.json missing in desktop app. - Check the Claude Code account, local tool state, and provider configuration involved in the failing workflow.
- Compare the failing environment with a known working setup, then change one configuration value at a time.
Platform/tool-specific checks
- Verify the command, editor, extension, or API client that produced the error.
- Compare local settings with CI, deployment, or editor-level settings when the error appears in only one environment.
- Avoid deleting credentials, local model data, or project settings until the failing scope is clear.
Step-by-step troubleshooting
- Capture the exact error message and the command, editor action, or request that triggered it.
- Check whether the failure is account/auth, quota/rate, model/provider, local runtime, or deployment configuration.
- Review the source evidence below and compare it with your environment.
- Apply one change at a time and rerun the smallest failing action.
- Keep the working fix documented for the team or deployment environment.
How to prevent it
- Keep provider/tool configuration documented.
- Record non-secret diagnostics such as tool version, provider name, model name, and command path.
- Add a lightweight check before CI or production workflows depend on the tool.