What this error means
MCP server failed to connect or authenticate (about:blank infinite loop) is a Claude Code failure pattern reported for developers trying to developer adds custom mcp server to claude code desktop/claude.ai but oauth flow never initiates — browser opens about:blank creating infinite loop. cli --transport http works but desktop/web gui fails.. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.
Why this happens
Anthropics/claude-code GitHub issue #11814 documents fully spec-compliant custom MCP OAuth servers failing in Claude Desktop and claude.ai web interface while working in CLI mode. Users see about:blank loop instead of OAuth consent page. Additional Asana MCP OAuth redirect_uri rejection (RFC 8252 loopback blocked) compounds the issue. Category: AI Coding Tools (Claude Code mapping per rules). High commercial value — breaks paid AI coding workflow with custom integrations.
Common causes
- Anthropics/claude-code GitHub issue #11814 documents fully spec-compliant custom MCP OAuth servers failing in Claude Desktop and claude.ai web interface while working in CLI mode. Users see about:blank loop instead of OAuth consent page. Additional Asana MCP OAuth redirect_uri rejection (RFC 8252 loopback blocked) compounds the issue. Category: AI Coding Tools (Claude Code mapping per rules). High commercial value — breaks paid AI coding workflow with custom integrations.
Quick fixes
- Confirm the exact error signature matches
MCP server failed to connect or authenticate (about:blank infinite loop). - Check the Claude Code account, local tool state, and provider configuration involved in the failing workflow.
- Verify the account session, API key, provider settings, and environment where the failing tool is running.
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.