From dec7599590a3dfa01402a5269de29c5dc1bcd235 Mon Sep 17 00:00:00 2001
From: spacedandy-agents-simplex
 <342056592+spacedandy-agents-simplex@users.noreply.github.com>
Date: Thu, 15 Oct 2026 12:21:26 +0000
Subject: [PATCH] ci: run the whole device test plan in both layouts with
 Claude [claude-test]

One Haiku 5.5 session per layout, one-hand UI first, each bounded by 700 turns, $10 and an auto-compact window of 200k tokens, so a long session does not re-read an ever larger context every turn. The validation session (run 38729704027) drove C2, D2 and D7 correctly in 55 turns for $0.03.
---
 scripts/ci/ios-sim/claude-scope   | 8 ++++----
 scripts/ci/ios-sim/claude-test.sh | 1 +
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/scripts/ci/ios-sim/claude-scope b/scripts/ci/ios-sim/claude-scope
index 12f5d1da96..bf87cbf922 100644
--- a/scripts/ci/ios-sim/claude-scope
+++ b/scripts/ci/ios-sim/claude-scope
@@ -1,6 +1,6 @@
 # What claude-test.sh runs: one Claude session per layout (reachable, top), the plan's cases, and each session's bounds.
 model: claude-haiku-5-5
-layouts: reachable
-cases: C2, D2, D7
-max-turns: 80
-max-budget-usd: 2
+layouts: reachable top
+cases: all of the plan's cases (C1-C9, D1-D15, U1-U3, N1-N11, W1-W11, G1-G4), skipping those named under "How to run the cases"
+max-turns: 700
+max-budget-usd: 10
diff --git a/scripts/ci/ios-sim/claude-test.sh b/scripts/ci/ios-sim/claude-test.sh
index e888652dac..1001308d50 100755
--- a/scripts/ci/ios-sim/claude-test.sh
+++ b/scripts/ci/ios-sim/claude-test.sh
@@ -90,6 +90,7 @@ for layout in $layouts; do
       --strict-mcp-config \
       --max-turns "$max_turns" \
       --max-budget-usd "$max_budget" \
+      --autocompact 200k \
       --no-session-persistence \
       --output-format stream-json \
       --verbose \