From a17edf43b7c0a94975fa4a00ad0706da7b6e7c6a Mon Sep 17 00:00:00 2001 From: ChemaVX Date: Thu, 13 Aug 2026 21:53:20 +0000 Subject: [PATCH] docs(short): un comprobador se puede derrotar por la forma de lo que comprueba Co-Authored-By: Claude Opus 5 --- docs/shortsmith-phases-4-6.md | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/docs/shortsmith-phases-4-6.md b/docs/shortsmith-phases-4-6.md index 1b2619e..33cf7f1 100644 --- a/docs/shortsmith-phases-4-6.md +++ b/docs/shortsmith-phases-4-6.md @@ -130,6 +130,21 @@ What actually landed, and where it differs from the plan below: the model verbatim and two wordings of one failure is how an error message stops being useful. +- **A check can be defeated by the shape of the thing it checks** (2026-08-13). The + grounding checker joins a `quote` list before looking for it, which is what closed the + Socorro hole in August: `“LIKE ALUMINUM` + `SMOOTH, NO WINDOWS”` join into one + sentence, no source contains it, rejected. But the join is defeated by giving each + line its own pair of quote marks — then they are two quotes, each grounded on its own, + and the spec passes in silence while the frame draws a sentence nobody said. Two of + the five Shorts generated that day had it. The rule now checks the *shape* rather than + the content — two opening marks are two quotes, whatever the sources say — and it runs + in `validate_spec`, so it costs a retry. It is a hard error and not an editorial note + on purpose: a fabricated quote attributed to a named witness is the worst failure this + system has, and a retry is cheap against it. Worth watching: given the choice between + picking a shorter verbatim span and dropping the quote marks, both rewrites dropped + the marks. Truthful, but a paraphrase is weaker than a quote — if that becomes the + habit, the fix is in the prompt, not the check. + Original plan, kept for the record: ### 4b (as planned). Narration (TTS) + burned-in captions