In remote mode an enterprise MCP config drops all other MCP servers, explaining each one.
What's wrong with this entry?
When a session runs in remote mode and an enterprise-managed MCP config file (managed-mcp.json) is present, MCP servers from every other source are dropped, each with a per-server explanation. Servers passed with --mcp-config get a matching warning instead of silently disappearing.
- Requires the
CLAUDE_CODE_REMOTEenvironment variable plus the presence of an enterprise managed-mcp.json. - Joins the existing hermetic-mode drop path; the shared drop helper gained a branch reporting the reason as "enterprise MCP config".
Ignored: an enterprise MCP config (managed-mcp.json) is present and has exclusive control over MCP servers
Strings lifted out of the shipped bundle, so the claim above can be checked against them.
Related
Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.
-
v2.1.246
Local file reads route through the storage backend when it serves the space
Both mention managed json
-
v2.1.227
Cloud gateway login URL is documented as admin-only
Both mention managed json
-
v2.1.242
Session spawn aborts on host-specific write paths committed to a repo
Both mention json