Why Cursor generates inefficient logic

1---
2description: Performance and algorithmic efficiency rules
3globs: "src/**/*.ts"
4alwaysApply: true
5---
6
7## Performance Rules
8- Use Map/Set for lookups instead of Array.find/Array.includes in loops
9- Use hash maps for pair/group finding (O(n) not O(n^2))
10- Avoid nested loops over the same or related collections
11- Use Set for deduplication instead of Array.filter with indexOf
12- Concatenate strings with Array.join, not += in loops
13- Pre-sort arrays when doing repeated binary searches
14- Use early returns/continue to reduce unnecessary iterations
15- Cache computed values that are used multiple times
16
17## Complexity Targets
18- Searching: O(log n) or O(1) with hash maps
19- Sorting: O(n log n) maximum
20- Filtering/mapping: O(n) single pass
21- Finding pairs/groups: O(n) with hash maps, not O(n^2) nested loops

1// BAD prompt (produces O(n^2)):
2// 'Find duplicate values in this array'
3
4// GOOD prompt (produces O(n)):
5// 'Find duplicate values in this array using a Set for
6// O(n) time complexity. Do not use nested loops.'
7
8// BAD prompt (produces O(n*m)):
9// 'Find common elements between two arrays'
10
11// GOOD prompt (produces O(n+m)):
12// 'Find common elements between two arrays in O(n+m)
13// time using a Set for the smaller array.'

1// Follow-up prompt after receiving generated code:
2// What is the time and space complexity of this function?
3// If it is worse than O(n log n), suggest an optimization
4// using appropriate data structures (Map, Set, sorting).

1// Cursor Chat prompt (Cmd+L, Ask mode):
2// @codebase Find all instances of these inefficient patterns:
3// 1. Array.find() or Array.includes() inside a loop
4// 2. Nested for/forEach loops over related collections
5// 3. String concatenation with += inside loops
6// 4. Array.indexOf() for deduplication
7// 5. Repeated .filter().map() chains that could be one pass
8// For each finding, show the efficient alternative.

1// BEFORE (O(n^2)):
2const duplicates = arr.filter((item, index) =>
3  arr.indexOf(item) !== index
4);
5
6// Select and press Cmd+K:
7// 'Optimize this deduplication to O(n) using a Set.'
8
9// AFTER (O(n)):
10const seen = new Set<number>();
11const duplicates: number[] = [];
12for (const item of arr) {
13  if (seen.has(item)) duplicates.push(item);
14  else seen.add(item);
15}