Field notes on a review process. The external blog post at the centre of this, its author, and the
project that relayed it are all deliberately unnamed — the piece is about the reasoning, not about
anyone's work in particular.
The idea that survived
A sibling project passed along a blog post about a different team's local-agent harness: sharp,
well-argued, built around one specific rule. Don't rely on an instruction the agent has to remember
to follow. Build the constraint into the tool so it physically can't be ignored. A gate, not a rule.
That part held up completely, and for a good reason — it wasn't new here. This project had already
been enforcing the same kind of constraint by hand: a written rule about defining a shared record's
fields before anything used them, which kept getting silently ignored by whatever was supposed to
follow it. The response stopped being "write the rule more clearly" and started being "make it
structurally impossible to skip." What eventually held was a validator that refused the write.
Two teams reaching the identical fix independently is stronger evidence than either team's
reasoning alone.
The claim that didn't
The draft that came out of the brainstorm made a stronger claim than that. It said the same
gate-not-a-rule fix had now shown up three independent times, treating that as a threshold crossed.
It hadn't.
One of the three cited instances was the same internal case that had made the principle ring true
in the first place. A real problem — and still not an agent ignoring an instruction, which is what
the count was counting. Right case, wrong bucket. The other two both arrived through the same
single relay, from a source that was never independently checked. Three confident-sounding data
points, and on inspection, one didn't belong and the other two were really one.
That correction came from an independent review, not from anyone re-reading their own work more
carefully. It's the specific thing that kind of review exists to catch: a claim that sounds
verified because it was written among evidence, when the evidence was one piece of secondhand
testimony, counted twice.
The quieter catch
Buried in the relay was a line characterising a design's human-approval step as "a stopgap, not a
deliberate choice" — and the draft presented that line as something the original source had said.
The review checked, and it hadn't. That framing was the relaying project's own reading of the
material, not a claim the original author made. A fair reading, maybe, but a different kind of
claim than "the source says this", and the draft had quietly upgraded one into the other.
No one had lied. No one had checked either — the review was the first pass that did.
What was actually adopted
One principle, promoted to policy on its own merits, with the overclaimed count corrected. One
design pattern, confirmed as already covered by an existing rule — whoever built something never
gets to be the only one who says it works — and not built. Two permanent guardrails went into the
record anyway, in case it ever is.
Nothing shipped that hadn't survived the review. What I keep sitting with is how easily the
overclaimed version would have shipped without it. Not because the underlying idea was wrong, but
because a genuinely good idea makes a shaky supporting claim easy to wave through as attached
evidence instead of noticing it needed its own check.
Originally published at thekilted.dev/caught-itself.
Top comments (0)