Codex MCP: Unable to connect to Context7 (Docker startup timeout / npx server fails)

What version of Codex is running?

0.80

What subscription do you have?

Bussiness

Which model were you using?

_No response_

What platform is your computer?

Darwin 25.2.0 arm64 arm

What issue are you seeing?

I’m unable to connect to the Context7 MCP server from Codex, using either:

1. the Docker-based server (mcp/context7), or
2. the documented npx -y @upstash/context7-mcp command.

Docker consistently times out during startup even after increasing timeouts. The npx approach fails immediately with a generic “Tip: Use /feedback to send logs to the maintainers when something looks off.” message.

What I tried

Option A — Docker (local)

Codex command

codex mcp add context7 -- docker run -i --rm mcp/context7

Config (~/.codex/config.toml)

[mcp_servers.context7]
command = "docker"
args = ["run", "-i", "--rm", "mcp/context7"]
startup_timeout_sec = 120
tool_timeout_sec = 120

Result
• Codex can’t connect; server startup hits a timeout even with startup_timeout_sec = 120.

Option B — Docs command (npx)

As documented here:

[https://developers.openai.com/codex/mcp#%3A~%3Atext%3Dcodex%20mcp%20add%20context7%20%2D%2D%20npx%20%2Dy%20%40upstash/context7%2Dmcp](url)

Command

codex mcp add context7 -- npx -y @upstash/context7-mcp

Result
• Codex shows a generic message: “Tip: Use /feedback to send logs to the maintainers when something looks off.” (no actionable details).

What steps can reproduce the bug?

Uploaded thread: 019bbbfc-e914-7321-a20b-3d9121a7de08

What is the expected behavior?

Codex should successfully start and connect to the Context7 MCP server using either Docker or the documented npx approach.