MCP tools/list different per user: what Sume's list varies by
MCP 2026-07-28 says list endpoints no longer vary per connection. Sume's tools/list still varies by credential: OAuth read, OAuth write or an API key.

Yes, a Sume tools/list differs by credential. The MCP 2026-07-28 changelog says list endpoints "no longer vary per-connection" because protocol sessions are removed, but Sume's list is built from the request's surface: an OAuth mcp:read session sees read-only tools, while mcp:write or an API key sees the full hosted set.
The two are not in conflict: the spec change is about per-connection state, and a credential is part of each request. Spec text is from the MCP changelog; Sume's from the docs and source, read 2026-10-01.
What did the spec change?
The changelog removes protocol-level sessions and the Mcp-Session-Id header, and says tools/list, resources/list and prompts/list no longer vary per connection. Servers needing cross-call state use explicit handles passed as ordinary tool arguments.
What decides which tools a Sume session sees?
| Session auth | What you see |
|---|---|
OAuth mcp:read only | Read-only tools; a mutating call returns insufficient_scope |
OAuth mcp:read + mcp:write | Full hosted tool set |
| API key | Full hosted tool set |
Why does a tool I expected not show up?
The docs say mutating and paid tools are hidden until the session has mcp:write or an API key. If a tool exists in the catalog but not in your list, check the scope first; see catalog shows a tool that is not in tools/list.
How should I cache the list?
Key any cache by credential, not just by URL. A read-only token and an API key hitting the same endpoint get different lists, so sharing one entry would show a tool the read-only token cannot call, or hide one the key can. Re-fetch after the scope changes, such as after the user grants write on consent.
Sources
Related posts
More in Developers
- MCP workload identity federation: Sume takes code grant only
MCP's roadmap names Workload Identity Federation. Sume's hosted MCP advertises only the authorization_code grant, so headless workloads use an API key.
- MCP x-mcp-header and Mcp-Param headers vs Sume idempotency_key
The MCP 2026-07-28 spec can mirror tool parameters into Mcp-Param headers via x-mcp-header. Sume takes idempotency_key as a tool argument in the JSON body.
- Ming Design-Layer splits a design into RGBA layers: Sume does not
Ming-Image-0.1-Design-Layer decomposes a flat design into RGBA PNG layers. Sume's Image API returns finished images, so build layers from transparent elements.
- Ming-Image-0.1-Design on Sume: not in the catalog, use these
Ming-Image-0.1-Design is an open 6B design model. Sume's image catalog does not list it, so send a listed model id such as openai/gpt-image-2.5 for posters.
Written by Sume