@@ -390,3 +390,24 @@ was not in judging them right or wrong — it was in separating the finding from
390390| --- | --- |
391391| #2813 | ` /v1/models ` and ` /api/models ` dumps from an account actually in Luna Reserve |
392392| #1419 | macOS ` .ips ` crash frames from a recurrence on Bun 1.4.0 |
393+ ### Why #1527 was opened and then put down
394+
395+ The fix looked ready: ` envelope_exhausted ` at
396+ ` src/adapters/cursor/protobuf-request.ts:1505 ` silently sets ` continuationMode = `
397+ ` "full-replay" ` , ` CursorRootEnvelopeLimitError ` already exists in ` cursor-errors.ts ` , and
398+ the file already imports it. Twenty lines, maybe.
399+
400+ Then I read the comment sitting directly under that assignment. It records that the
401+ reason is deliberately NOT propagated to the checkpoint store, that writing it there was
402+ MEASURED inert because ` live-transport.ts ` prepares a spread copy, that reaching the store
403+ needs the reason threaded through ` PreparedCursorRunRequest ` , and that this is a
404+ signature change on the shared prepare path which "belongs to its own phase". It cites
405+ the audit rounds that established each of those.
406+
407+ Someone already stood where I was standing, went further than I had, and wrote down why
408+ they stopped. Adding a throw on top of that without re-deriving their measurements would
409+ not be finishing their work — it would be overwriting a conclusion I had not earned. The
410+ cheap version of this fix is exactly the version the comment warns against.
411+
412+ So #1527 stays open with its design recorded in ` 001 ` and this note attached. It is not
413+ blocked on missing information; it is blocked on deserving the change.
0 commit comments