Record four standing decisions, including one unrepairable binding

REPO-020a: CL-20260802-018 was swept into checkpoint 92f1eed with no Change-Log
trailer, so git log --grep for it returns nothing, and adding the trailer would
mean rewriting a released commit. The exception is now recorded in two places
deliberately -- the CHANGELOG preamble's Commit-binding section, so a reader
learning the general rule learns it has an exception, and the entry's own field,
so a reader arriving there directly need not know the preamble exists. This
commit also names CL-20260802-018 in prose, so an un-anchored grep resolves to
something pointing at 92f1eed, which the trailer grep never can.

REPO-024: the vendored frontend-design skill has not drifted. SHA-256 matches on
both files against the cached marketplace copy -- a byte comparison, not a
reading.

REPO-037: the adversarial worktree-clean check stays removed. checkSimplification
still performs no rewrite, no ticket implements one, and restoring an uncalled
function to serve a caller nobody has scheduled is not an improvement. Whichever
ticket adds the rewrite must bring the check back with it.

REPO-038: the bootstrap field stays. My own ticket text said the server
populates a field nothing reads, and that was wrong -- three test files outside
web/client assert on its presence and absence as a live proxy for whether a
session was established. Removing it from the client alone would have created an
inconsistency for no gain. Whoever revisits it should replace the proxy with a
direct assertion rather than delete the signal.

Committed with --no-verify at the user's explicit request.

TODO: REPO-020a, REPO-024, REPO-037, REPO-038

Change-Log: CL-20260802-066
Dev-Log: DL-20260802-076
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: claude

CHANGELOG

@@ -50,6 +50,36 @@ Dev-Log:
 
 Entries
 -------
+Change-ID: CL-20260802-066
+Commit: pending
+Date: 2026-08-02
+Type: Record -- four standing decisions closed
+Request-or-TODO: REPO-020a, REPO-024, REPO-037, REPO-038
+Outcome: REPO-020a records the one entry whose commit binding cannot be
+  repaired. CL-20260802-018 was swept into checkpoint 92f1eed with no
+  Change-Log trailer, so git log --grep for it returns nothing and adding the
+  trailer would mean rewriting a released commit. The exception is now written
+  in two places -- this file's Commit-binding preamble, so a reader learning the
+  general rule learns it has one exception, and the entry's own field, so a
+  reader arriving there directly does not have to know the preamble exists.
+  REPO-024: the vendored frontend-design skill is checked and has not drifted --
+  SHA-256 matches on both files against the cached marketplace copy.
+  REPO-037: the adversarial worktree-clean check stays removed. checkSimplification
+  still performs no rewrite, no ticket implements one, and restoring an uncalled
+  function to serve a caller nobody has scheduled is not an improvement.
+  REPO-038: the bootstrap field stays. It is dead in web/client, but three test
+  files outside it assert on its presence and absence as a live proxy for
+  whether a session was established
+Affected-behavior: None. Records and decisions only
+Compatibility-or-migration: None
+Verification: The drift check is a byte comparison, not a reading. The
+  REPO-038 finding is three cited test files, not an absence of grep hits
+Known-limitations: REPO-038's proxy is still a proxy -- whoever revisits it
+  should replace it with a direct assertion rather than delete the signal.
+  This commit names CL-20260802-018 in prose so an un-anchored grep resolves to
+  something pointing at 92f1eed, which the trailer grep never can
+Dev-Log: DL-20260802-076
+
 Change-ID: CL-20260802-065
 Commit: pending
 Date: 2026-08-02

DEVLOG

@@ -25,6 +25,31 @@ Next-safe-step:
 
 Entries
 -------
+Dev-Log: DL-20260802-076
+Date: 2026-08-02
+Status: Four decisions recorded
+Change-ID: CL-20260802-066
+Commit: pending
+Request-or-TODO: REPO-020a, REPO-024, REPO-037, REPO-038
+Goal: Close four standing records rather than leave them as open questions
+Assumptions: A decision nobody wrote down is not a decision
+Decisions: Recorded the CL-20260802-018 exception in two places deliberately.
+  One would have been enough to be true and not enough to be found: a reader who
+  arrives at the entry directly has no reason to consult the preamble, and a
+  reader learning the general binding rule deserves to know it has an exception.
+  Kept the bootstrap field. My own ticket text said the server populates a field
+  nothing reads, and that was wrong -- three test files outside web/client
+  assert on it, using its presence and absence as a proxy for whether a session
+  exists. Removing it from the client alone would have created an inconsistency
+  for no gain
+Files-or-schemas: CHANGELOG preamble and the CL-20260802-018 entry
+Validation: The skill drift check is a byte comparison; the bootstrap finding
+  cites three specific test files rather than resting on an absence of hits
+Failures-or-discarded-approaches: Removing the bootstrap field, abandoned once
+  its real readers were found
+Known-limitations: REPO-038's proxy remains a proxy
+Next-safe-step: None; all four are closed
+
 Dev-Log: DL-20260802-075
 Date: 2026-08-02
 Status: Wired and browser-driven against a fixture