What this error means
MCP OAuth completes but token is not honored — server stays in 'Needs authentication' is a Claude Code failure pattern reported for developers trying to fix oauth 2.1/pkce auth flow for remote mcp servers where claude code reports success but tools never become available. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.
Why this happens
GitHub Issue anthropics/claude-code#60260 (open, created 2026-05-18, 3 comments). Has repro label, area:auth + area:mcp labels. OAuth flow succeeds in browser, curl bypass works perfectly, only Claude Code bridging fails. High commercial value — developers using corporate MCP tools with Claude Code Enterprise/Claude Pro cannot access any MCP tools. Category mapping: Claude Code → AI Coding Tools per SKILL.md category table.
Common causes
- GitHub Issue anthropics/claude-code#60260 (open, created 2026-05-18, 3 comments). Has repro label, area:auth + area:mcp labels. OAuth flow succeeds in browser, curl bypass works perfectly, only Claude Code bridging fails. High commercial value — developers using corporate MCP tools with Claude Code Enterprise/Claude Pro cannot access any MCP tools. Category mapping: Claude Code → AI Coding Tools per SKILL.md category table.
Quick fixes
- Confirm the exact error signature matches
MCP OAuth completes but token is not honored — server stays in 'Needs authentication'. - 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.