Repository navigation
Unclear purpose or source of the repository.id property #173
Description
Activity
The intent is this:
Or owned by the source forge (e.g. github)?
We want to store an
id, not just a repository name, because it is actually quite common for MCP servers to move repository names (the common case is that someone starts an MCP server as a personal project and then "promotes" it to their organization, thus changing the GitHub owner, but an operation which should retain the GitHubid).PR to improve documentation around this would be very welcome 🙏
PR to improve documentation around this would be very welcome 🙏
- added a commit that references this issue
on Jul 14, 2025 - added a commit that references this issue
on Aug 7, 2025 - added a commit that references this issue
on Sep 12, 2025
Describe the bug
In the OpenAPI YAML there is a repository.id required property. It is required in the server.json schema too.
Is this generated by the MCP registry? Or owned by the source forge (e.g. github)? I am guessing either this is generated by the MCP registry, in which case it should not be required in the server.json schema, or it is defined by the source forge and recorded here to prevent resurrection attacks.
The example values look kind of like UUIDs but GitHub has numeric or GraphQL style IDs. I am not sure what to make of it.
The publisher tool does not require it:
registry/tools/publisher/server.json
Lines 50 to 53 in 6a2e5d1
But maybe that is an oversight or because the tool can compute it using the GH context akin to:
To Reproduce
The YAML OpenAPI schema has this, but no description:
registry/docs/openapi.yaml
Lines 86 to 91 in 6a2e5d1
The JSON schema has this, but no description:
registry/docs/server-json/schema.json
Lines 6 to 12 in 6a2e5d1
Expected behavior
I would like to understand!