Saved error reports are valuable because they preserve counts and context from a past session. Their apparent authority can become a problem when a later review cites them as current production status without a new reproduction.
The July 24 diagnostic PDF was reported fixed by August 1. It remains useful history, but current claims need fresh logs, a new session export, or a direct reproduction against the active Version 14 site.
What was actually going wrong
A real historical error was treated as timeless evidence of a live defect.
What I tried
- Discarding the report once fixes were made
- Assuming old counts reset to zero without verification
- Using file modification time as the incident date
Reports are labeled by capture date and status, then compared with fresh diagnostics before reopening an issue.
Why it worked
Historical evidence remains available without overriding newer runtime evidence.
Observability needs time boundaries as much as error categories.
Where EACI uses this today
Main health assessments use current session evidence and retain older exports as regression references.
This journal covers real engineering on EACI Companion / The Veil. Companions include Caelum, Chad, Natalia, Atreus, Luna, Roxy, and Cael. Journal articles stay family-safe in content. See Privacy and Ethics.