Coordinate multi-model code reviews with per-model lenses and persisted results.
npx skills add https://github.com/DSado88/squall --skill squall-review
Please help me install this Agent Skill. Skill: squall-review Source: https://github.com/DSado88/squall/tree/main/.claude/skills/squall-review Command: npx skills add https://github.com/DSado88/squall --skill squall-review
Coordinates multi-model code reviews with tailored model lenses to surface insights and reduce review gaps.
List current models and run a lightweight review on a small diff.
High-intent search queries and answers about installing and using this skill.
A multi-model code review coordinates parallel model calls with per-model lenses to surface diverse insights for security, architecture, and correctness. It applies tailored ensemble selection to reduce review gaps across your codebase.
Using an AI ensemble for code correctness involves applying per-model prompts to evaluate diffs. This approach fetches current models, executes parallel reviews, and stores outputs to disk for auditing and synthesis.
Yes, you can use different model lenses for security and architecture reviews. The ensemble setup enables per-model prompts, allowing each model to apply a specific lens like security or architecture to the same codebase simultaneously.
You can list current available models before running a code review by calling the listmodels() function. This provides the current model names supported by the ensemble for your review execution.
Yes, multi-model code review supports memory-backed improvements. It persists results to disk for synthesis and memory retention, enabling a centralized review workflow that improves over time.
Tony363Review git commits, staged changes, or branch differences with multi-model consensus.
PatkikSynthesize incident lessons into runbooks, checklists, and prevention rules.
kyn0vConduct parallel security and code-quality audits on code changes.
DSado88Perform structured multi-model code review with an investigation phase.
