phase: 06_system_page_gpu_verdicts

All criteria verified and the project is fully green.

**Verified (final pass — all 3 phase-06 tasks were already in `complete/`)**
- `classify("gpu.busy_pct", 99)` / `gpu.mem_used_pct` → `None`; `THRESHOLDS` has no `gpu.*` row; `alerts.py`/`api.py` carry no GPU verdict; live browser: deskwork GPU at **91 %** and **98 %** renders `class="hero"`, `color=rgb(22,25,31)` (plain text), no bar, no `.sev-*`.
- GPU temp on the hero `.sub` from live agents: `Radeon 8060S Graphics · VRAM 165.1 MiB / 512.0 MiB · pool 78% · 48 °C`, R9700 `· 40 °C`, gateway prints none (no `n/a`, no `0 °C`).
- Charts collapsed by default: fresh load = 3 panels / 3 toggles / **0 open / 0 `<svg>`**; expand → 3 SVGs (510 px wide), repaints across polls (legend 3.3 %→5.6 %), collapsed siblings keep their own height (803 vs 358/339).
- `gpu.busy_pct@0` still answers from `/api/system/history` (345 points) — verdict gone, data intact.

**Tests / lint / coverage**
- `bash .agents/validate.sh` → **EXIT=0**; `uv run pytest --cov --cov-fail-under=94` → **550 passed, 8 skipped, 97.91 %**; `uv run ruff check src tests agent` → clean; `uv run pyright` → 0 errors; `node --check` all scripts clean; `uv pip list --outdated` empty.
- `uv run pytest -m live --live` → **8 passed** (real agents/engines/upsd); dev server restarted, `/api/health` all sources `ok`.
- New screenshots: `data/logs/phase06-final-collapsed-default.png`, `data/logs/phase06-final-expanded-gpu-busy-uncoloured.png`.

**Completion criteria:** classify→None ✅ · pytest+94 % gate ✅ · ruff+pyright ✅ · validate.sh ✅ · screenshots of the three behaviours on live data ✅ · AGENTS.md severity section claims no GPU rows ✅

**Docs:** fixed 3 items — (1) `AGENTS.md` pytest count stale (546 → 550); (2) `app.js` `severityFor` docstring still claimed "`page-ups.js` still calls it (phase 05 converts it)", contradicting the code/tests/AGENTS.md (comment-only); rest of README/AGENTS reviewed in full (env table vs `config.py`+`.env.example`, agent routes incl. `/schema`, alerts rules, page descriptions, contracts) — consistent.

**Notable:** fixed a real pre-existing defect in `scripts/dev-server.sh` — `stop` slept 1 s and deleted the pidfile without waiting for exit, so `restart` with a browser open overlapped two servers on one SQLite file, producing a `database is locked` store refusal that `/system` rendered as a red event. Now waits (bounded 20 s) then SIGKILLs; verified a restart with live polling logs no such error/event. No live infra config was changed.

**Next pending phase:** `.agents/phases/todo/07_llm_monitor_style_pages`

tests/test_alerts.py

@@ -216,21 +216,28 @@ def test_a_number_the_upstream_does_not_publish_is_never_a_breach(
     assert alerts.build(SETTINGS, [state()], [payload]) == []
 
 
-def test_a_gpu_breach_names_the_card_and_dedupes_per_metric(
-        host_scrape: Any, agent_snapshot: dict[str, Any]) -> None:
+def test_a_saturated_gpu_is_not_an_alert(host_scrape: Any,
+                                         agent_snapshot: dict[str, Any]) -> None:
+    """Phase 06: these are LLM servers — a fully-worked GPU is the job, not a fault.
+
+    The host is otherwise all-clear, so a card at 100 % busy with both memory pools
+    filled must produce **no** alert at all: `severity.THRESHOLDS` has no `gpu.*` row,
+    the payload carries no GPU verdict, and this module owns no threshold of its own
+    from which to resurrect one.
+    """
     def fill_the_card(snap: dict[str, Any]) -> None:
         gpu = snap["gpus"][0]
-        gpu["busy_pct"] = 97.0
+        gpu["busy_pct"] = 100.0
         for used, total in (("vram_used_bytes", "vram_total_bytes"),
                             ("gtt_used_bytes", "gtt_total_bytes")):
             if gpu.get(total) is not None:
                 gpu[used] = gpu[total]
 
     payload = host_scrape(agent_snapshot, cpu_busy=10.0, mutate=fill_the_card)
-    messages = {item["id"]: item["message"] for item in alerts.build(SETTINGS, [state()], [payload])}
-    assert any("GPU utilisation 97% on Radeon AI PRO R9700" in text
-               for text in messages.values()), messages
-    assert any("GPU memory" in text and "R9700" in text for text in messages.values())
+    assert payload["gpus"][0]["busy_pct"] == 100.0, "the card really is saturated here"
+    assert alerts.build(SETTINGS, [state()], [payload]) == [], (
+        "a busy GPU produced an alert — GPU verdicts were removed in phase 06"
+    )
 
 
 def test_a_degraded_agent_warns_and_names_the_sections_that_failed(