Adjacent findings — real, actionable, not threats¶
The pass will surface things that are worth acting on and are not threats. A contract clause that does not cover a task that will certainly happen. A document describing an architecture the design has moved away from. A scope decision nobody has taken.
Forcing these into the findings register corrupts it – a reader is entitled to assume every
F-line describes a way the system can harm someone or be harmed. Dropping them loses work the pass paid for. So they land here, asADJ-n, a separate series.
What belongs here¶
| Contract and scope | A task the work requires that the agreements do not cleanly authorise |
| Document drift | Two documents describing the same thing differently, where one is stale |
| Spec gaps | A design decision the pass assumed was made, and finds was not |
What does not belong here. Anything describing a way the system can harm someone or be harmed. That is a finding, however awkward its category.
How these resolve¶
Not by being fixed here. Each entry names where it goes to die – a document to amend, a decision to take, a conversation to have. An entry with no route out is not finished.
ADJ-1 · ¶
Raised \<date>, \<cell> · Status open · Route out \<the specific thing that closes this>
What happens. \<the situation, plainly>
Why it is a question. \<what is ambiguous or missing, with references>
Why it matters. \<the consequence of leaving it>
Options, recorded rather than decided.
| \<option> | \<what it costs> |
| \<recommended option> | Recommended. \<why> |