What this error means
MCP OAuth calls fail when coordinator agent dispatches subagent via agent_toolset_20260401 — vault_ids parameter not available in subagent context is a Anthropic API failure pattern reported for developers trying to fix mcp oauth authentication failure for dynamics 365 / microsoft graph mcp servers called from subagents dispatched via claude managed agents. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.
Why this happens
Anthropic anthropic-sdk-python #1562 (2026-05-18): vault_ids configured on sessions.create not passed to subagent execution context. Any MCP tool call requiring OAuth credentials fails silently or returns auth error. Category: Anthropic API (Claude platform issue affecting paid managed agents feature). Distinct from any covered error.
Common causes
- Anthropic anthropic-sdk-python #1562 (2026-05-18): vault_ids configured on sessions.create not passed to subagent execution context. Any MCP tool call requiring OAuth credentials fails silently or returns auth error. Category: Anthropic API (Claude platform issue affecting paid managed agents feature). Distinct from any covered error.
Quick fixes
- Confirm the exact error signature matches
MCP OAuth calls fail when coordinator agent dispatches subagent via agent_toolset_20260401 — vault_ids parameter not available in subagent context. - Check the Anthropic API 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.