Find code blind spots with independent reviews.
What it does
Finds problems in code changes through independent model reviews and returns a synthesized verdict without automatically applying fixes.
When to use it
Use when asked for an adversarial or multi-model review, a code stress test, or help finding blind spots.
How to use it
Give it code changes to challenge; it returns a synthesized review without automatically applying changes.
What you provide
- An existing project
Good to know
- starts one helper agent per item, so the count depends on the work
- edits its own instructions
Latest review
What you need · 1
Requires its Subagents guidance to select each reviewer's reasoning level and speed; the instruction provides no alternative guidance.
Access · 2
This skill
a separate pull request
Write
Opens a model-default update PR
~/.cursor/rules/pstack-models.mdc
Read
Reads your reviewer model settings