Review code diffs for hot-path complexity, allocations, and N+1 patterns.
npx skills add https://github.com/dhnt/dhnt --skill performance-review-dhnt
Please help me install this Agent Skill. Skill: performance-review Source: https://github.com/dhnt/dhnt/tree/main/catalog/md/review/performance-review Command: npx skills add https://github.com/dhnt/dhnt --skill performance-review-dhnt
Prevents slowdowns by guiding reviewers to identify performance regressions caused by changes in algorithms, allocations, hot-path behavior, or runtime characteristics.
Ask an AI to review a PR’s diff for performance regressions and return a structured report with a summary, prioritized concerns/hardening opportunities, and benchmark/profiling recommendations.
High-intent search queries and answers about installing and using this skill.
To detect performance regressions in a pull request, review the diff for hot-path complexity, new allocations, and N+1 patterns. This process identifies runtime risks in frequently executed logic before code ships.
A code performance review analyzes changes in algorithms, allocations, and hot-path behavior to prevent slowdowns. It looks for resizing, string work, and per-item operations inside frequently executed code paths.
You identify N+1 patterns and allocation behavior by reviewing your diff for per-item DB, RPC, or file operations introduced within loops. This scrutiny surfaces runtime risks in request-handling logic.
Performance review relies on measurement-driven validation, meaning benchmarking and profiling are required to confirm impact. It avoids speculative claims by producing targeted benchmark guidance alongside prioritized concerns.
Yes, you can use this performance review for pre-merge gating when latency, throughput, or memory usage may have changed. It provides a structured summary distinguishing hardening opportunities from actual concerns.
The best way to audit algorithmic complexity is reviewing code changes for hot-path behavior and runtime characteristics. This approach detects increased complexity and validates impact using profiling rather than speculation.
StartaleGroupAnalyze pull request diffs to assign review tiers and agents.
frankxaiProfile CPU, memory, I/O, and tracing to identify and eliminate performance bottlenecks.
arcasilesgroupScan codebases for performance anti-patterns across multiple languages.
martin-hsu-testMonitor a URL for 10 minutes post-deploy to detect console errors, latency regressions, and page failures.
