How do I verify Montage workspace access with GitHub Copilot Chat?
GitHub Copilot Chat supports MCP in several IDEs, but their setup and prerequisites differ. This guide concerns the Copilot extension in JetBrains IDEs, rather than duplicating a VS Code connection recipe. Organization policy and the extension's installed version must both permit MCP. A Copilot subscription, an IDE login, and an authorized Montage workspace are three separate conditions, not interchangeable evidence of readiness.
Using GitHub Copilot Chat, call only Montage signals_status as the first live operation. Report the actual workspace identity and coverage from the tool response. If authentication, transport, or the tool is unavailable, stop and explain the blocker. Do not start collection, recurring tracking, writes, publishing, messaging, or social actions.Select the JetBrains documentation tab
Use the JetBrains IDEs tab in GitHub's official Copilot Chat MCP documentation and match its prerequisites to your installed extension. Do not apply the VS Code 1.99 requirement or .vscode configuration instructions blindly to JetBrains. Check the extension's MCP controls before editing files; the IDE's separate JetBrains AI Assistant integration is not the same client or permission context.
Check managed MCP policy
For an organization-provided Copilot Business or Enterprise subscription, confirm the MCP servers in Copilot policy permits use. GitHub documents that managed policy separately from personal-plan access. A forbidden server should remain blocked rather than being hidden inside a different tool entry. Keep the intended Montage workspace explicit so approval of an editor feature does not become approval of a data-sharing scope.
Configure a custom remote server
Follow the JetBrains extension's manual MCP registration path for the installed release and supply https://app.montages.co/mcp as the hosted URL. The registry route only applies to entries actually listed there; this guide makes no registry or verified-connector claim for Montage. Preserve other server definitions and avoid command-launcher examples that would start a local process instead of reaching the hosted service.
Resolve consent and tool availability
Complete only the remote authentication method supported by that Copilot extension release. If Montage OAuth cannot be completed, report the limitation rather than copying authentication behavior from VS Code. Review which tools become available in the active chat and keep writes unapproved. Discovery is useful for checking schema availability but is not evidence that this account can execute workspace operations.
Verify the intended workspace from chat
Ask the JetBrains Copilot chat to execute only signals_status on Montage. Inspect the returned identity and coverage before asking for research, and keep the result distinct from any model-written summary. A rejected call should end setup until its cause is resolved. Do not authorize collection, recurring tracking, or social actions through a broad editor trust prompt or a successful tool listing.
Keep the boundaries clear.
- This guide targets the GitHub Copilot extension in JetBrains IDEs, not VS Code or JetBrains AI Assistant.
- Documented setup, not live interoperability certification. A visible tool catalog is not workspace authorization.
- First live operation: read-only signals_status. No automatic collection, recurring tracking, or social actions; broader actions need separate explicit authorization.
Start with a read-only check: ask your agent to report actual source coverage before collecting records or exporting an audience.
Primary references
Documentation reviewed 2026-11-15. Product capabilities, client availability and plans can change. Third-party marks identify their products, not an affiliation.