Stage 372: PR #2407 — docs: advance runtime adapter gate by @Michaelyklam

CHANGELOG.md

@@ -2,6 +2,10 @@
 
 ## [Unreleased]
 
+### Documentation
+
+- Document the #1925 runtime-adapter gate update: Slice 1 run-journal replay has now passed a 100-trial synthetic replay/restart validation pass on current `origin/master`, #2313's selected-session chat SSE cap is shipped, and Slice 2 is ready for a reversible adapter-seam planning PR without moving execution ownership yet.
+
 ## [v0.51.78] — 2026-05-16 — Release BB (stage-371 — stuck-PR sweep salvage — RTL chat + ambient quota chip with composer-clutter gate)
 
 ### Added

docs/rfcs/hermes-run-adapter-contract.md

@@ -4,7 +4,7 @@
 - **Author:** @Michaelyklam
 - **Updated by:** @franksong2702
 - **Created:** 2026-05-11
-- **Revised:** 2026-05-14
+- **Revised:** 2026-05-16
 - **Tracking issue:** [#1925](https://github.com/nesquena/hermes-webui/issues/1925)
 
 ## Credit and Scope
@@ -49,6 +49,29 @@ The immediate goal is not to build a sidecar. The immediate goal is to define th
 browser contract, classify current runtime state, and gate the first reversible
 journal slice.
 
+## Current Gate State — 2026-05-16
+
+Slice 1 is now past the first active validation gate:
+
+- #2283 shipped the run-journal replay layer in v0.51.71.
+- A 100-trial synthetic replay/restart validation pass against current
+  `origin/master` passed on 2026-05-16. The matrix covered completed-run replay,
+  interrupted stale-pending recovery, fresh-pending grace handling, StreamChannel
+  reconnect ordering, duplicate-prevention merge behavior, many-session recovery,
+  large-journal derivation, and stream-to-turn-id lifecycle linking.
+- The focused regression set
+  `tests/test_turn_journal.py tests/test_turn_journal_lifecycle.py tests/test_stale_stream_pending_recovery.py`
+  also passed on the same worktree.
+- #2393, shipped through v0.51.76, capped live chat token SSE transports to the
+  selected conversation pane. Background sessions now rely on existing
+  status/replay/reattach behavior instead of keeping one live `/api/chat/stream`
+  EventSource per active session.
+
+This evidence does not prove the future runner/sidecar path. It does mean the
+project should stop treating Slice 1 as purely passive observation and can move to
+Slice 2 planning: introduce the adapter seam over the still-legacy journaled path
+without moving execution ownership yet.
+
 ## Goals
 
 - Preserve the current rich WebUI workbench experience.
@@ -269,6 +292,10 @@ Success criterion:
 
 ### Slice 2: Adapter interface over the journaled legacy path
 
+Status as of 2026-05-16: ready for a planning/adapter-seam PR after the active
+Slice 1 validation pass and the #2313 selected-session stream cap. Slice 2 should
+still be a reversible boundary change, not a sidecar or execution-ownership move.
+
 Scope:
 
 - introduce the `RuntimeAdapter` interface only after Slice 1 proves replay,