Adjudicate backlog: drop finished Phase 2 and Phase 4, narrow Phase 3

Phase 2 (both lines) — done. The headless review spawns the enabled
subagents and prints each verdict under its own heading (src/bot.py),
proven end-to-end on ten31-site PR #4; the `agents +/-` handler updates
and atomically persists each room's enabled set. Finished milestone, and
the record already lives in AGENTS.md Decisions + Current state.

Phase 4 (Spark Control tile) — already built elsewhere. spark-control's
Bots tab has had a first-class gitea-review-bot entry since v0.28.0, same
code path as matrix-bridge; the running dashboard reports it live. The
work was never this repo's. Left a one-line pointer under Deferred.

Phase 3, deploy-on-a-non-static-repo — dropped. No repo needs it (only
the static ten31-site is onboarded), and it isn't doable as written: the
deploy step is hardcoded to the static-site publish script and the only
per-repo knob is on/off, not a per-repo deploy command. Exercising it
means a real deploy plus a force-merge against a live service. Revisit
when a concrete non-static repo needs it — that repo is the honest test
target, and the deploy-command field comes first.

Phase 3, three-repo onboarding — kept, annotated with its check plan and
a correction: reviews run serially, so the stated "no interleaving" exit
is trivially satisfied. What matters at 3+ repos is a long review
head-of-line-blocking the next room's poll.

New: heartbeat for wedge detection. Scoped to the gap the tile leaves —
it reads container state, so it catches a crash but not a stuck sync/poll
loop, which looks identical to healthy from outside.

ROADMAP.md

@@ -3,21 +3,20 @@
 Phases beyond the first milestone (threaded single-flow review + merge/reject/deploy; see AGENTS.md
 `## Current state`).
 
-## Phase 2 — Subagent panel, proven
-- The lead `claude -p` reliably spawns the enabled subagents (reviewer / adjudicator /
-  security-auditor) and presents each opinion + its own recommendation. **Exit:** a real PR review
-  shows up to 3 distinct, labeled opinions + the lead's call; flaky headless subagent spawning
-  falsifies it → fall back to bot-orchestrated separate `claude -p` runs (the rejected alternative).
-- In-chat `agents +/-` toggles change which subagents run on the next review, per room (persisted).
-
 ## Phase 3 — Multi-repo at scale
 - Onboard ≥3 repos to their own rooms; confirm independent polling, no cross-talk. **Exit:** three
-  repos reviewed from three rooms in one run with no interleaving.
+  repos reviewed from three rooms in one run. *(adjudicated 2026-07-18: verify — needs no new code;
-- Per-repo deploy (`[repo.<name>].deploy_on_merge`) exercised on a non-static repo (not a Pages site).
+  rooms are already independent (per-room heads/threads/agents are namespaced, merge-confirm state is
-
+  keyed by room+PR). Check plan, when a real 2nd/3rd repo is onboarded: watch one poll cycle in
-## Phase 4 — Spark Control tile
+  `docker compose logs -f`. Note the exit criterion is mis-stated — reviews run **serially** (one async
-- Status badge + Update/Restart/Stop-Start/Logs, mirroring matrix-bridge's D10 (captured in the
+  task walks the rooms in turn), so "no interleaving" is trivially satisfied; the thing actually worth
-  cross-project inbox as a spark-control item).
+  watching at 3+ repos is a long review **head-of-line-blocking** the next room's poll, i.e. the latency
+  problem the deferred poll→webhook item anticipates.)*
+- Emit a heartbeat the poll loop touches, so a **wedged** bot (container up, sync/poll loop silently
+  stuck) is distinguishable from a healthy one — today it looks identical from outside. Crash detection
+  is already covered by spark-control's tile, which reads the container's docker state; this is only the
+  gap that leaves. Lightest correct form is a heartbeat file a Docker healthcheck or any monitor can
+  read, **not** an HTTP status server (this repo has no HTTP surface; don't add one for a badge).
 
 ## Deferred / non-goals
 - Not a Claude-session bot (that's matrix-bridge); not Maubot (revisit at ~6+ bots or web-UI mgmt);
@@ -25,3 +24,5 @@ Phases beyond the first milestone (threaded single-flow review + merge/reject/de
 - A shared "bot kit" library (extract the common matrix-nio + SSH plumbing across the three bots)
   once duplication bites — the lighter step before ever considering Maubot.
 - Poll → webhook upgrade if 60s latency ever matters.
+- Ops/dashboard tile lives in **spark-control**, not here — its Bots tab has shipped a gitea-review-bot
+  tile since v0.28.0 (status badge, Update, Start/Restart/Stop, View logs), confirmed live.