Replies: 2 comments
-
|
I've seen this happen because The CLI performs a capability check against your organization's settings the moment you use a prompt ( To fix this, you'll likely need an admin to toggle that policy to "Enabled" or "Unverified" in the GitHub web UI. IIRC, the CLI currently treats the agentic harness initialization as a "third-party" request context in some environments, so it hits that wall before it even processes your local flags. If you can't change the Org policy, check if you can run the command without the prompt flag to see if it's purely the agentic path causing the trigger, though for your use case that might not be an option. |
Beta Was this translation helpful? Give feedback.
-
|
@Nickalus12
Could you clarify what exactly “third-party MCP servers” refers to in this context? I need to explain this to our organization admin before requesting a policy change. I need to explain this clearly to our organization admin before requesting a policy change.If there’s any official documentation describing this, it would be very helpful if you could share a URL as well. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
I am using the GitHub Copilot CLI in GitHub Actions.
Even though I have added the disable-builtin-mcps option, I still get the following error:
What could be causing this?
The workflow YAML is defined as follows:
Beta Was this translation helpful? Give feedback.
All reactions