fix(tests): key the corpus-column scratch dir per call, not per process

bootstrap/tests/corpus_zig_bodies.rs

@@ -83,7 +83,12 @@ fn per_spec_carries_three_zig_digits_and_says_so_in_its_header() {
         eprintln!("zig not on PATH -- SKIPPED");
         return;
     }
-    let dir = std::env::temp_dir().join(format!("t27-corpuscols-{}", std::process::id()));
+    // Unique per CALL, not per process: a pid is one value for the whole test
+    // binary, so every test in this file would resolve the same directory and
+    // the `remove_dir_all` below would erase a sibling's input mid-read.
+    static NTH: std::sync::atomic::AtomicUsize = std::sync::atomic::AtomicUsize::new(0);
+    let nth = NTH.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
+    let dir = std::env::temp_dir().join(format!("t27-corpuscols-{}-{nth}", std::process::id()));
     std::fs::create_dir_all(&dir).expect("temp dir");
     let path = dir.join("per-spec.txt");
     let out = Command::new(env!("CARGO_BIN_EXE_t27c"))

docs/now/2026-08-31-the-scratch-gate-went-red-on-master-the-hour-it-was-wired.md

@@ -0,0 +1,8 @@
+# NOW -- The scratch gate went red on master the hour it was wired (2026-08-31)
+
+## The scratch gate went red on master the hour it was wired (Closes #2968)
+
+- tri harness scratch --gate has been red on master since 994f3c8a3: bootstrap/tests/corpus_zig_bodies.rs keys its scratch directory by std::process::id(), which is one value for the whole test binary.
+- Measured, and stated as measured: only one of that file's three tests creates or deletes the directory today, so nothing collides on this tree. The gate is structural, and right to be -- the second test to use that key would collide, and would pass the first time it ran.
+- Fixed as the gate's own message prescribes: an AtomicUsize counter, unique per call. Gate now prints none and exits 0; the three tests still pass.
+- The gate found this within an hour of being wired, on a file written after it. That is the reading worth keeping: it went red on new code rather than on history.