Skip to content

Commit c49fed6

Browse files
authored
Merge pull request #1976 from lidge-jun/codex/promote-main-w5d
Promote dev to main: alert-precision record
2 parents 286cbd2 + 7cf8587 commit c49fed6

1 file changed

Lines changed: 45 additions & 7 deletions

File tree

‎devlog/_plan/260817_wave5_execution/090_wave6_closeout.md‎

Lines changed: 45 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -243,12 +243,22 @@ targeting the default branch. Every campaign PR targets `dev`, so none of them c
243243
CodeQL feedback at all — verified: `refs/pull/1959/head` and `refs/pull/1963/head` have analyses
244244
because they target `main`, while the `preview`-targeting pair have zero.
245245

246-
The actual root cause is duller and more useful: **`dev` is scanned on push, and nobody read
247-
the result.** It carries 84 open alerts against `main`'s 71. The finding was observable on the
248-
integration branch from the moment #1897 merged until promotion, and no step in this campaign
249-
ever looked. CodeQL is structurally absent from the `dev` PR flow *and* the post-merge alert
250-
list is not part of anyone's gate — those are two different holes, and I named neither the
251-
first time.
246+
**Second attempt, also wrong.** I then wrote that `dev` is scanned on push and nobody read the
247+
result. Both halves are false: `dev`'s most recent analysis is `02abe0afa` from 2026-08-15,
248+
default setup runs on a *weekly* schedule, and `0be660a2e` is not an ancestor of that commit —
249+
so the code was never in a `dev` scan at all. Its 84 alerts are stale, not current, which is the
250+
opposite of what I claimed they showed. Alert #87's only instances are `refs/heads/main` and
251+
`refs/pull/1959/head`. "Nobody read it" described a page that never displayed it.
252+
253+
**The actual missed signal, third time.** `github-advanced-security[bot]` posted the finding as
254+
an inline review comment on **#1959** at `src/providers/antigravity-models.ts:273` at
255+
**02:38:08Z**. #1963 promoted at **02:55:04Z**. It was sitting on a promotion PR, in the review
256+
thread, for **17 minutes** before the code reached `main` — and I was actively editing that PR's
257+
description during the window. Not a coverage gap. I did not look at the review comments on a
258+
PR I was in the middle of rewriting.
259+
260+
Three explanations for one mistake, the first two of which blamed infrastructure. The third is
261+
the one that is true and the least comfortable.
252262

253263
Severity in context: the input is a configured `baseUrl`, so exploitation needs a hostile or
254264
careless config rather than attacker-controlled traffic. Worth fixing, not urgent. Separately,
@@ -267,7 +277,7 @@ edge cases (empty string, all-slashes, no trailing slash, interior slashes).
267277

268278
**The root cause is a process one and belongs in the record.** #1897 merged on local focused
269279
tests plus `tsc`. That substitutes for CI on the axis it covers — behavior — and silently skips
270-
the axis it does not: static analysis. Waiting for full CI would have surfaced this before it
280+
the axis it does not: static analysis. ~~Waiting for full CI would have surfaced this before it~~
271281
reached `main`. The instruction for this run was to stop waiting on per-PR CI and gate once at
272282
the end, which is a reasonable trade for speed; the honest accounting is that it traded away
273283
exactly this class of finding, and the end-gate I ran (`bun test`, `typecheck`, `privacy:scan`)
@@ -295,3 +305,31 @@ returns true. The alert is disclosed on #1897, on both promotion PR pairs, and h
295305
What I did not do, and stand by: I never approved a promotion PR. `MAINTAINERS.md` forbids
296306
authors approving their own, and the rulesets require a code-owner review. That the promotion
297307
happened by another route is the maintainer's call to make, not mine to route around.
308+
### Alert #87: verified fixed on the code, not yet flipped by GitHub
309+
310+
Stating this precisely, because "fixed" and "closed" are different claims and only one of them
311+
is currently provable.
312+
313+
**The code is fixed and promoted.** `59d57a9bf` is an ancestor of `dev`, `preview` and `main`,
314+
and all three trees are byte-identical. A reviewer fuzzed 400,000 adversarial strings — slashes,
315+
`\u2028`, lone surrogates, NUL — against the replaced regex and found **zero** behavioral
316+
differences.
317+
318+
**The alert still reads `open`.** That is scan lag, not a live finding: the most recent
319+
JavaScript/TypeScript analysis on `main` ran at `7979903b9`, which predates the fix. Queried
320+
against `refs/pull/1968/head` — the branch that *does* contain it — alert #87 returns **zero**.
321+
So the fix is confirmed by scan, just not yet on the `main` ref. It should flip on the next
322+
JS/TS run there; until it does, this campaign does not claim it closed.
323+
324+
### What I did not fix, and should say so
325+
326+
A reviewer asked whether other instances of the same pattern remain. **Yes — 30 occurrences of
327+
`/\/+$/` across `src/`**, with open `js/polynomial-redos` alerts on at least six
328+
(#83, #84, #60, #53, #51, #52) covering `openai-chat-url.ts`, `openai-responses-url.ts`,
329+
`openai-responses.ts`, `openai-chat.ts` and `anthropic.ts`.
330+
331+
All predate this campaign (created 2026-08-12/13), so the scoped claim — *this campaign
332+
introduced exactly one and fixed exactly that one* — is accurate. But they take the same
333+
attacker-influenceable `baseUrl` input my own fix comment argues about, so leaving them
334+
unmentioned would be the convenient framing rather than the honest one. They are out of scope
335+
here and worth their own pass.

0 commit comments

Comments
 (0)