fix(bin/wire-company-brain.sh): heredoc inside $(...) breaks bash 3.2

CHANGELOG.md

@@ -16,6 +16,7 @@ three commits past it), and a bug report can name a release instead of a sha nob
 Sections dated before 2026-09-19 predate the cycle and stay as they are.
 
 ## Unreleased
+- fix(bin/wire-company-brain.sh): **a heredoc built inside `$(cat <<EOF ...)` does not parse under bash 3.2** — the `$(...)` scanner reads the heredoc body as shell text and an apostrophe in it ("company's") opens a quote it never closes, so `doctor.sh` reported the script unparseable on macOS's stock bash while bash 5 here parsed it fine. Rebuilt with `IFS= read -r -d '' block <<EOF` (no heredoc inside `$(...)`), byte-identical output verified. The same trap was live in two more places, `claude-hooks/hooks/context-ceiling.sh` and `context-watch.sh` (a heredoc feeding python's stdin inside `$(...)`), fixed the same way with output re-verified against a synthetic transcript. Added a `bin/check-portability.sh` check for the pattern (excludes `<<<` here-strings and `$((...))` arithmetic shift, which false-triggered the raw grep) — workshop field report, macOS bash 3.2
 - learn(bug-logs): **7 new drafts, 3 addenda and one feature-ask note from a module-rename field run, each re-checked on v0.24.0 where a probe was cheap.** New: `rename module` leaves XPath naming the old module; `rename page` renames a same-named folder instead (reproduced; root cause is a name match with no type check); a case-only `move` leaves an empty folder twin; a quoted DataGrid 2 association path passes `check` and fails CE1613; `check --references` ignores an in-script `rename module`; `diff-local` misses new (untracked) units; and `diff-local` prints a UUID for the module. Addenda: BUG-62 (no `rename snippet`), BUG-63 (`microflow_type` case, SQL-vs-Starlark entity types, `activity_count` scope; title widened), BUG-102 (legacy grid XPath has no offline route). Paste-ready upstream drafts for four, not filed. — field report from a colleague's naming-conventions project (mxcli v0.23.0, Mendix 11.12.4)
 - learn(skills/learned-mdl-preflight.md): **STOP row 26: DataGrid 2 column and `sort by` paths over an association.** Write the column path unquoted, and do not sort over an association on ≤ v0.24.0. Both wrong forms pass `check --references` and fail CE1613 at build. — field report from a colleague's naming-conventions project (mxcli v0.23.0, Mendix 11.12.4)
 - learn(skills/lint-that-actually-runs.md): **three more lint vocabulary facts, verified on v0.24.0, with a before/after rule.** `microflow_type` is `MICROFLOW`/`NANOFLOW`/`RULE`, and `microflows()` yields all three. SQL `PERSISTENT` is Starlark `Persistent`. `activity_count` counts top-level objects (splits, loops, annotations), not actions, and `activities_for()` does not see loop bodies. — field report from a colleague's naming-conventions project (mxcli v0.23.0, Mendix 11.12.4)

bin/check-portability.sh

@@ -121,6 +121,22 @@ while IFS= read -r f; do
           'Use cd "$(dirname "$x")" && pwd.' ;;
     esac
 
+    # bash 3.2's $(...) scanner does not recognise a heredoc opened inside it: it scans
+    # the body as shell text while hunting for the matching ')', so a stray apostrophe or
+    # quote in the body breaks the parse on macOS's default bash — invisible to bash -n on
+    # any machine that only has bash 5 (real incident: bin/wire-company-brain.sh, a
+    # "company's" in the block text). Strip the two lookalikes first: a `<<<` here-string
+    # (no heredoc body, just overlaps `<<`) and `$((...))` arithmetic (its `<<` is a
+    # bit-shift) — both false-triggered the raw pattern in testing.
+    case "$txt" in
+      *'$('*'<<'*)
+        _pc="$(printf '%s' "$txt" | sed -e 's/<<<//g' -e 's/\$(([^)]*))//g')"
+        if printf '%s' "$_pc" | grep -Eq '\$\([^)]*<<[^<]'; then
+          report "$f" "$ln" "heredoc opened inside \$(...) — unparseable on bash 3.2" \
+            "Read the heredoc into a variable first (IFS= read -r -d '' var <<EOF ... EOF), then use \"\$var\" inside \$(...)."
+        fi ;;
+    esac
+
     case "$txt" in
       *'mapfile '*|*'readarray '*)
         report "$f" "$ln" "mapfile/readarray need bash 4 (macOS ships 3.2)" \

bin/wire-company-brain.sh

@@ -43,7 +43,12 @@ $extra"
 done
 
 BEGIN='<!-- COMPANY-BRAIN:BEGIN -->'; END='<!-- COMPANY-BRAIN:END -->'
-block="$(cat <<BLK
+# bash 3.2's $(...) scanner does not recognise a heredoc opened inside a command
+# substitution: it scans the body as shell text, so an apostrophe in the body (e.g.
+# "company's") opens a quote that is never closed and the parse fails with "unexpected
+# EOF while looking for matching `''" on macOS's default bash. Build the block with a
+# heredoc feeding `read -d ''` instead, which every caller supports the same way.
+IFS= read -r -d '' block <<BLK || true
 $BEGIN
 ## Company brain
 
@@ -53,7 +58,7 @@ integration, and before choosing a component, read \`$BRAIN/ROUTING.md\`** and f
 that fires. It loads on demand; nothing from it is copied here.
 $END
 BLK
-)"
+block="${block%$'\n'}"
 written=0
 while IFS= read -r CL; do
   [ -n "$CL" ] || continue

claude-hooks/hooks/context-ceiling.sh

@@ -88,7 +88,11 @@ if [ -f "$cache" ]; then
 fi
 
 if [ -z "$ctx" ]; then
-  ctx=$("$PY" - "$transcript" <<'PY' 2>/dev/null
+  # bash 3.2 does not recognise a heredoc opened inside $(...): it scans the body as
+  # shell text while hunting for the matching ')', so a stray quote in the body can
+  # break the parse on macOS's default bash (same trap as bin/wire-company-brain.sh).
+  # Read the heredoc into a variable first, then pipe it in — no heredoc inside $(...).
+  IFS= read -r -d '' _ctx_py <<'PY' || true
 import json, os, sys
 path = sys.argv[1]; CAP = 8 << 20
 def newest(chunk):
@@ -116,7 +120,7 @@ try:
         else: print(0)
 except Exception: print(0)
 PY
-)
+  ctx=$(printf '%s' "$_ctx_py" | "$PY" - "$transcript" 2>/dev/null)
   case "$ctx" in ''|*[!0-9]*) ctx=0 ;; esac
   printf '%s %s' "$now" "$ctx" > "$cache" 2>/dev/null
 fi

claude-hooks/hooks/context-watch.sh

@@ -69,7 +69,11 @@ printf '%s' "$now" > "$throttle" 2>/dev/null
 # here can be 600KB (an inlined screenshot), so a fixed byte tail routinely
 # slices mid-line and finds no parseable record. Stops at the first usage
 # record found, or at the cap.
-ctx=$("$PY" - "$transcript" <<'PY' 2>/dev/null
+# bash 3.2 does not recognise a heredoc opened inside $(...): it scans the body as
+# shell text while hunting for the matching ')', so a stray quote in the body can
+# break the parse on macOS's default bash (same trap as bin/wire-company-brain.sh).
+# Read the heredoc into a variable first, then pipe it in — no heredoc inside $(...).
+IFS= read -r -d '' _ctx_py <<'PY' || true
 import json, os, sys
 
 path = sys.argv[1]
@@ -111,7 +115,7 @@ try:
 except Exception:
     print(0)
 PY
-)
+ctx=$(printf '%s' "$_ctx_py" | "$PY" - "$transcript" 2>/dev/null)
 
 case "$ctx" in ''|*[!0-9]*) exit 0 ;; esac
 [ "$ctx" -gt 0 ] || exit 0