2026-09-27 17:06:40 +00:00
1 changed files with 10 additions and 1 deletions
@@ -242,7 +242,16 @@ pays for itself furthest away.
      `x-event-id`. Claims checked against a running server are marked *verified* in the text, so
      a reader can tell what was measured from what was reasoned. One limitation lifts with the
      transport: a module may now call another's tool, which issue 049 recorded it could not.
- [ ] 3.4 the controller's link on NATS
+- [~] 3.4 the controller's link on NATS — **the seam exists and the outbound half is through
+      it.** `Bus` is stated in the mesh's words (publish an event, declare to a node) rather than
+      a transport's, with an AMQP and a NATS implementation, both shipping: steps 1 to 4 leave
+      every node on AMQP, and both shipping is what lets one conformance fixture hold them to the
+      same envelope. The NATS one is checked against a real server, reading back from the stream
+      rather than from the code that wrote it.
+
+      The seam turned out to be eight call sites — the same smallness that said this bus could be
+      replaced at all. Still on `*amqp.Channel`: `RequestBuild` and `Ask`, which carry reply-queue
+      machinery, and the whole consume side (the control loop, enrolment, serving).
 - [ ] 3.5 the host's link on NATS — mirroring, still importing nothing
 - [x] 3.6 the tool runtime's client on NATS, behind the unchanged sdk contract — round-tripped
      against a real server: a tool answered across two connections, a throwing handler reaching