Custom MCP tools start successfully and are selected, but are unavailable in Copilot automations

[normura]

🏷️ Discussion Type

Bug

💬 Feature/Topic Area

Copilot Cloud Agent

Body

Summary

My custom Azure DevOps MCP server works in manual Copilot cloud sessions, but its tools appear unavailable to the agent in sessions started by an automation.

The automation's environment setup reports that the MCP server started successfully and lists all 10 configured tools. All 10 tools are also explicitly selected in the automation editor. However, the agent cannot invoke them.

Configuration

• Surface: GitHub Copilot cloud automations
• MCP server: ado, running @azure-devops/mcp as a local process
• Authentication: PAT authentication through an Agents secret
• Model selection: Auto
• Actual model: gpt-6-luna in both the manual and automation sessions
• MCP server startup log reports version 2.11.0

The same MCP configuration and credentials work in manual cloud sessions.

Steps to reproduce

1. Configure the Azure DevOps MCP server in the repository's Copilot MCP settings.

2. Verify that core_list_projects works in a manual cloud session.

3. Create an automation in the same repository.

4. Explicitly select the custom MCP tools in the automation's Tools selector.

5. Use this prompt and run the automation:

   Use the ado MCP server to invoke core_list_projects with top set to 1. If tool discovery is required and available, use it first. Report whether an actual tool call occurred and its result. Do not substitute shell commands or direct HTTP requests.

6. Observe that environment setup starts ado successfully and lists its tools, but the agent does not invoke an ADO tool.

Creating a brand-new automation produces the same behavior.

Expected behavior

The automation agent can invoke the selected custom MCP tools, as it can in a manual cloud session.

Actual behavior

The startup log states:

MCP server started successfully (version 2.11.0) with 10 tools

It lists:

ado/core_list_project_teams
ado/core_list_projects
ado/work
ado/wit_work_item
ado/wit_query
ado/wit_work_item_attachment_upload
ado/wit_work_item_attachment_link
ado/wit_work_item_write
ado/wit_work_item_link_write
ado/search_workitem

When asked to list its callable tools, the agent reports the seven GitHub MCP tools selected in the automation and several built-in tools, but none of the selected ADO tools.

It also reports that no tool-search or deferred-tool discovery mechanism is available.

No ADO tool call occurs, so this run produces no Azure DevOps authentication or connectivity error.

The callable-tool inventory is the agent's self-report. I do not have the underlying model-request diagnostics to independently verify which tool definitions were supplied.

Screenshots

Attached:

• Successful MCP startup with all 10 tools listed + agent response reporting that the ADO tools are unavailable:

• Automation editor showing all 10 ADO tools selected:

• Successful MCP startup, tool discovery and call in a manual cloud session:

Questions

Are custom MCP tools supported in Copilot automations?

If so, is there an additional permission or configuration requirement beyond repository MCP configuration and selecting the tools in the automation editor?

Is there a way to inspect the actual tools exposed to the model in an automation session?

[github-actions[bot]]

💬 Your Product Feedback Has Been Submitted 🎉

Thank you for taking the time to share your insights with us! Your feedback is invaluable as we build a better GitHub experience for all our users.

Here's what you can expect moving forward ⏩

• Your input will be carefully reviewed and cataloged by members of our product teams. 
  • Due to the high volume of submissions, we may not always be able to provide individual responses.
  • Rest assured, your feedback will help chart our course for product improvements.
• Other users may engage with your post, sharing their own perspectives or experiences. 
• GitHub staff may reach out for further clarification or insight. 
  • We may 'Answer' your discussion if there is a current solution, workaround, or roadmap/changelog post related to the feedback.

Where to look to see what's shipping 👀

• Read the Changelog for real-time updates on the latest GitHub features, enhancements, and calls for feedback.
• Explore our Product Roadmap, which details upcoming major releases and initiatives.

What you can do in the meantime 💻

• Upvote and comment on other user feedback Discussions that resonate with you.
• Add more information at any point! Useful details include: use cases, relevant labels, desired outcomes, and any accompanying screenshots.

As a member of the GitHub community, your participation is essential. While we can't promise that every suggestion will be implemented, we want to emphasize that your feedback is instrumental in guiding our decisions and priorities.

Thank you once again for your contribution to making GitHub even better! We're grateful for your ongoing support and collaboration in shaping the future of our platform. ⭐

[tjhiggins]

Same issue trying to use the sentry mcp in an automation

[WuGing]

Same experience with sentry and azure.