code-optimizer

You can audit codebases for performance anti-patterns by spawning 13 parallel specialist agents that use pattern-based detection via Grep and Glob, analyzing references first to identify inefficiencies without reading source files upfront, reducing anchoring bias and speeding up optimization.

Parallel code optimization agents detect N+1 queries, memory leaks, unnecessary allocations, blocking I/O, concurrency issues, and large bundle sizes across multi-language codebases, reporting findings with severity, file location, explanation, fix, and estimated impact.

Pattern-based code review uses parallel specialist agents that load detection patterns from references and scan using Grep and Glob, avoiding the upfront code reading bias of manual audits and delivering faster, targeted optimization insights across large repositories.

Yes, parallel specialist agents apply to multi-language codebases across repositories, scanning for memory leaks, concurrency issues, and other performance problems using pattern-based detection before inspecting surrounding context to confirm findings.

The best way to identify N+1 queries and excessive allocations is running a parallel multi-agent audit that spawns 13 specialized agents to scan for different performance problem classes using pattern-based detection, generating actionable reports with estimated impact.

You generate actionable optimization reports by running parallel specialist agents that detect performance anti-patterns and report findings with file, pattern, severity, current snippet, explanation, fix, and estimated impact for each identified inefficiency.