Where the workflow shifted
Model routing is an operating model, not a marketing slogan.
Primary, secondary, fallback models should each have clear task boundaries.
Tool names are not outcomes
The signal matters when it changes how a team ships, reviews, or recovers work, not when it only names another tool.
Check permissions and failure
- Map task classes to primary/secondary/fallback models and define output format requirements
- Keep the test narrow: one low-risk task or tool entry before connecting permissions, logs, failure handling, and human takeover to production
What still needs proof
No fallback leads to unstable production behavior during model incidents. Keep the original source open so the announcement, the evidence, and this site's interpretation stay separate.