eltmon

code-review-performance

Deep performance analysis focusing on algorithms and resources

eltmon 31 1 Updated 3mo ago
GitHub

Install

npx skillscat add eltmon/overdeck/code-review-performance

Install via the SkillsCat registry.

SKILL.md

Performance Review

Algorithm Complexity

  • Identify O(n^2) or worse algorithms
  • Look for unnecessary iterations
  • Check for opportunities to use better data structures

Database/API Patterns

  • N+1 query detection
  • Missing indexes (check query patterns)
  • Unbounded queries (missing LIMIT)
  • Connection pool exhaustion risks

Memory & Resources

  • Memory leaks (unclosed resources)
  • Unbounded caches or buffers
  • Large object allocations in loops

Concurrency

  • Lock contention hotspots
  • Blocking operations in async contexts
  • Thread pool exhaustion

Blocking Operations (CRITICAL for Node.js)

  • execSync/spawnSync - These BLOCK the event loop and cause UI freezes
  • Search for: execSync, spawnSync, readFileSync in hot paths
  • ALWAYS flag as HIGH IMPACT - causes perceived hangs and latency spikes

Fix pattern to recommend:

// Replace execSync with async version
import { exec } from 'child_process';
import { promisify } from 'util';
const execAsync = promisify(exec);
const { stdout } = await execAsync('command here');

Common offenders in this codebase:

  • tmux capture-pane / tmux send-keys / tmux has-session
  • git branch / git status / git log

Output Format

For each finding:

  • Impact: High/Medium/Low
  • Scale factor: "At 10x load, this will..."
  • Location: file:line
  • Suggested optimization