diff --git a/FOLLOW_UP_2.md b/FOLLOW_UP_2.md
index 8ed0068..e853d51 100644
--- a/FOLLOW_UP_2.md
+++ b/FOLLOW_UP_2.md
@@ -17,48 +17,31 @@ The OperationGuard ordering fix in `handlers.rs` is the structurally most import
   replaces its UI operation map from that snapshot, and the React game-list
   merge uses the snapshot to clear stale download/install/update/uninstall
   spinners even when a begin/finish/fail event was missed.
+- Install-recovery matrix: `recovery_covers_install_matrix_rows` now covers the
+  ten named Installing / Updating / Uninstalling rows from PLAN.md, including
+  markerless reserved directories while an intent proves ownership.
+- Update rollback on commit-rename failure:
+  `update_commit_rename_failure_restores_previous_local` forces a post-extract
+  commit conflict without production hooks and verifies backup restore.
+- Uninstall delete-failure restore:
+  `uninstall_delete_failure_restores_backup` covers the rollback path on Unix,
+  where directory permissions can force recursive delete failure without hooks.
+- Installed-only → Ready rescan:
+  `rescan_promotes_installed_only_game_to_ready_when_sentinel_appears` pins the
+  cached-index transition after a user adds `version.ini` to a local-only root.
+- Scanner gating dispatch: local monitor tests now cover active-operation event
+  drops, burst collapse, fallback sideload pickup, and non-catalog roots leaving
+  no library entry or delta.
+- Serve gating dispatch: stream tests now cover `GetGame` response gates and
+  the shared full-file/chunk transfer gate for non-catalog, active-operation,
+  missing-sentinel, and local-path cases.
+- TempDir test helper: peer tests now share `test_support::TempDir` instead of
+  carrying per-module copies.
 
 ## Still open
 
-### Tests still missing
-
-#### 2. Install-recovery matrix (~10 rows)
-
-Only two intent-driven rows are covered: `Updating | no local | .local.installing | .local.backup` (`recovery_restores_backup_for_interrupted_update`) and the active-skip case (`startup_recovery_skips_active_game_roots`). The other ~9 combinations of `(intent_state ∈ {Installing, Updating, Uninstalling}, local present?, .local.installing present?, .local.backup present?)` are unverified.
-
-**Approach:** convert `recovery_restores_backup_for_interrupted_update` into a table-driven test. Iterate over the named matrix rows in PLAN.md, set up FS + intent, run `recover_game_root`, assert resulting FS + intent.
-
-#### 3. Update rollback on commit-rename failure
-
-Only extract-failure is tested (`update_failure_restores_previous_local`). The branch where extract succeeds but the staging→local rename fails is unexercised. Force it by making `local/` un-renamable (e.g. swap in a directory under that name post-extract) or by injecting a rename hook.
-
-#### 4. Uninstall delete-failure restore
-
-`restore_backup` rollback in `transaction.rs:107-114` is unreached by tests.
-
-#### 5. Installed-only → Ready rescan
-
-PLAN listed this. No test exercises the transition where a user drops `version.ini` into a `local/`-only game root and the next rescan flips it to Ready.
-
-#### 6. Scanner gating dispatch
-
-`handle_watch_event` and `RescanGate` / `run_gated_rescan` (`local_monitor.rs:208-287`) have zero tests. Cover:
-
-- Event for ID under active operation is dropped.
-- Burst of events collapses to ≤2 rescans for the same ID.
-- Sideload picked up by fallback scan.
-- Non-catalog game produces no library index entry and no `LibraryDelta`.
-
-#### 7. Serve gating dispatch
-
-`local_download_available` predicate is tested; `handle_get_game_command`, `handle_get_game_file_data`, `handle_get_game_file_chunk` dispatch paths aren't. Small tests against an in-memory `Ctx` covering: non-catalog ID, mid-operation, missing sentinel.
-
 ### Code hygiene
 
-#### 8. Consolidate `TempDir` test helper (getting worse)
-
-Reimplemented in **five** test modules now: `install/intent.rs`, `install/transaction.rs`, `local_games.rs`, `download.rs`, `handlers.rs`. Each round of new tests adds another copy. Move to a single `test_support` module (or use the `tempfile` crate, already stable). The longer this waits, the bigger the eventual consolidation diff.
-
 #### 9. `availability` field is `String`, not a typed enum
 
 `lanspread-db::Game::availability: String` with `AVAILABILITY_*` consts works but allows invalid values. `remote_peer::summary_from_game` already has a defensive fallback for unknown strings. A typed enum + serde would be tighter, but is a wire-format change — coordinate if/when other peers are upgraded.