Authentication for ACP over Streamable HTTP #2680
Replies: 1 comment
|
You could borrow the discovery half of MCP's authorization spec almost as is, since it already answers "what does this server accept". The agent serves Two side benefits. Agents that do want a full OAuth flow can point I work at Permit. This walkthrough goes through the same discovery chain on the MCP side (the 401, the protected resource metadata, then the authorization server metadata) step by step: https://www.permit.io/blog/oauth-on-mcp |
Uh oh!
There was an error while loading. Please reload this page.
The Streamable HTTP transport makes remote agents practical, and the RFD leaves auth "layered on top". That works for one deployment, but every agent ends up picking its own scheme. Goose's
serveuses a fixed shared secret inX-Secret-Keyor?token=, and a host that dials several agents has to carry each one's scheme.We dial agents over HTTP from a host that already signs short-lived tokens for its other services. Each connection carries a bearer JWT that the host signs, with an audience for the agent pool, checked against a published key set. Keys rotate by
kidand nothing secret is shared with the agent. It's the resource-server pattern MCP's authorization spec already uses.Could the transport say how an agent declares what it accepts? For example, a bearer token in
Authorization, plus where to find the issuer's keys, so a host knows what to send without reading each agent's docs. This sits next to the connection-scope point on #1977: if model credentials become per connection, the connection itself has to be authenticated first, and both should use the same mechanism.All reactions