Practices
Cross-model review
Having a different model, with a fresh context, review an agent's change before it merges.
TRIALOn the radar since 12 Oct 2026Draft
Why Trial
A second model with a fresh, read-only context reviews risky changes before they ship. The author's model never reviews its own risky work.
Movement
On other radars
No community radar places this yet.
Sources
- Claude Code documents subagents that run with their own context and model. checked 12 Oct 2026
Share and embed
Makers: show where your product sits. The badge updates when the radar does.
Share this entry
Also in Practices · Trial
- Compilers as agent feedbackChoosing languages and checks that catch agent mistakes before a human reviews the work.
- Mask personal data before a model sees itReplacing personal identifiers with placeholders before text reaches a language model.
- Models select, code verifiesLetting a model choose from source material while code checks the choice against the source.