diff --git a/plans/abstractions.md b/plans/abstractions.md
index 256c2a97..dd42c272 100644
--- a/plans/abstractions.md
+++ b/plans/abstractions.md
@@ -12,6 +12,12 @@ Anything short → _Watching_ (what's missing) or _Rejected_ (why).
 
 ## Last scan
 
+- **Date:** 2026-06-07 (radar loop, pass 31)
+- **Pass 31 — A1 conformance loop WORKLIST COMPLETE.** tcl excluded (foreign `*.tcl`); final:
+  4 migrated (common-lisp/erlang/feed/go) + 5 excluded (forth/js/ocaml/smalltalk/tcl). A1 =
+  **12 on shared driver + 6 excluded**; only the parity-gated merge to architecture remains.
+  commerce shipped a refund saga on flow (2nd flow use) + finished Phase 5 → going quiescent.
+  relations building graph algos (all-paths) — still unconsumed (W9 unchanged).
 - **Date:** 2026-06-07 (radar loop, pass 30)
 - **Pass 30:** conformance loop near done — `ocaml` + `smalltalk` excluded (both foreign
   `test.sh`/corpus runners, as predicted). Tally: 4 migrated, 4 excluded, **tcl only** left.
@@ -261,14 +267,21 @@ one merged file copied N times. Correct one-liner:
   prolog, search — `events` migrated via its OWN loop; `search` via counters mode (which
   corrects the earlier "search excluded" note). **+4 on the `loops/conformance` branch:
   `common-lisp` 487/487, `erlang` 761/761, `feed` 189/189, `go` 609/609** — pending merge.
-  **4 EXCLUDED — all foreign-runner harnesses** (correctly, not force-migrated): `forth`
-  (Hayes core.fr via awk+python), `js` (test262 `.js`/`.expected` fixtures), `ocaml`
-  (scrapes its `test.sh` + foreign `.ml` baseline), `smalltalk` (scrapes `test.sh` + `*.st`
-  corpus). Only **`tcl`** remains for the loop. **Emerging final shape: ~half the 9
-  candidates migrate (SX-suite harnesses), ~half are foreign runners excluded** — the
-  classify-then-exclude gate did exactly its job. After tcl, the loop's worklist is done;
-  the last A1 step is **merging `loops/conformance` → architecture under the adopter-parity
-  check** (the coordination flag above).
+  **5 EXCLUDED — all foreign-runner harnesses** (correctly, not force-migrated): `forth`
+  (Hayes core.fr via awk+python), `js` (test262 `.js`/`.expected`), `ocaml` (scrapes
+  `test.sh` + `.ml` baseline), `smalltalk` (scrapes `test.sh` + `*.st` corpus), `tcl`
+  (foreign `*.tcl` vs `# expected:` annotations).
+- **✅ CONFORMANCE LOOP WORKLIST COMPLETE (pass 31).** Final A1 picture:
+  - **12 on the shared driver:** acl, apl, datalog, events, haskell, mod, prolog, search
+    (on architecture) + common-lisp, erlang, feed, go (on `loops/conformance`, pending merge).
+  - **6 correctly excluded** (foreign-program runners — testing a language impl against an
+    external corpus is legitimately a different harness): forth, js, ocaml, smalltalk, tcl, lua.
+  - **Honest finding:** the driver's reach is narrower than the raw "15 conformance.sh"
+    count implied — language substrates that run real `.lua/.st/.ml/.tcl/.js/.fr` programs
+    *should* keep their foreign runners. ~half migrate, ~half don't, and that's correct.
+  - **One step left:** merge `loops/conformance` → architecture under the **adopter-parity
+    check** (the coordination flag above — the shared `lib/guest` driver change must be
+    proven non-regressive against all existing adopters first). The loop is now idle.
 - **NOW IN PROGRESS — dedicated loop (2026-06-07).** A human-triggered `conformance` loop
   (worktree `/root/rose-ash-loops/conformance`, branch `loops/conformance`, tmux session
   `a1-conformance`, briefing `plans/agent-briefings/conformance-loop.md`) is working the