This repository was archived by the owner on Jan 30, 2026. It is now read-only.
Repository navigation
This repository was archived by the owner on Jan 30, 2026. It is now read-only.
OpenAI Agent should share what MCP servers are worked or not on start up #400
Description
Prerequisites
- Write a descriptive title.
- Make sure you are able to repro it on the latest version
- Search the existing issues.
Steps to reproduce
- Configure MCP server tools
Expected behavior
on start up see that x MCP server are configured but also which servers are failed or not.Actual behavior
I can only see which mcp servers are ready to use with using the /mcpError details
n/aEnvironment data
Name Value
---- -----
PSVersion 7.5.2
PSEdition Core
GitCommitId 7.5.2
OS Microsoft Windows 10.0.26100
Platform Win32NT
PSCompatibleVersions {1.0, 2.0, 3.0, 4.0…}
PSRemotingProtocolVersion 2.3
SerializationVersion 1.1.0.1
WSManStackVersion 3.0Version
Preview 6
Visuals
On start up i see 3 MCP servers configured
BUT only two of them are actually ready to use.
We should have on start up "2 of 3 MCP servers ready to use"
Activity
daxian-dbw commentedon Jul 30, 2025
It takes time to start an MCP server. It could be a few seconds or sometimes even longer. So, we cannot start MCP servers in sequence but have to do it in parallel and to not block the startup of the AIShell. Today, we have no async mechanism for us to report MCP server failures on the go (like a pop-up window in UI). This is why today a user needs to run
/mcpto see the status of MCP servers.