Cursor rule

1---
2description: When doing code reviews
3alwaysApply: false
4version: 1.1.0
5---
6 
7# Code Review Standards
8 
9This rule defines when bot feedback is incorrect given context bots lack. Use this to
10identify suggestions that don't apply, not to skip valid feedback based on priority.
11 
12## Core Philosophy
13 
14Address all suggestions where the bot's analysis is correct given full context. Decline
15when you can articulate why the bot's reasoning doesn't hold - valid declines explain
16why the analysis is incorrect, not why addressing it is inconvenient.
17 
18## When Bot Suggestions Don't Apply
19 
20These patterns describe situations where bot analysis is typically incorrect. Decline
21with explanation when you can demonstrate the bot's reasoning doesn't hold.
22 
23### Single-Use Values
24 
25Bots flag inline values as "magic strings" needing extraction. This suggestion is wrong
26when the value appears exactly once and context makes the meaning clear. Extracting
27`METHOD_INITIALIZE = "initialize"` for a single use adds indirection without DRY
28benefit. Constants exist to stay DRY across multiple uses, not to avoid inline values.
29 
30### Theoretical Race Conditions
31 
32Bots flag potential race conditions based on static analysis. This suggestion is wrong
33when operations are already serialized by a queue, mutex, or transaction the bot can't
34see. Add synchronization when profiling or testing reveals actual race conditions.
35 
36### Redundant Type Safety
37 
38Bots suggest stricter types or null checks. This suggestion is wrong when runtime
39validation already handles the case correctly, or when the type system guarantees the
40condition can't occur. TypeScript serves the code - working code with runtime safety
41takes priority over compile-time type perfection.
42 
43### Premature Optimization
44 
45Bots flag performance concerns without data. This suggestion is wrong when no profiling
46shows actual performance problems. Optimize based on measurements - complexity should
47yield measurable performance gains.
48 
49## Items Requiring Case-by-Case Judgment
50 
51These require evaluation in context - sometimes the bot is right, sometimes wrong.
52 
53### Test Coverage Gaps
54 
55Bot requests for edge case tests: Address if the edge case could reasonably occur and
56cause user-facing issues. Decline if you can demonstrate the scenario is already handled
57by other validation or genuinely can't occur given system constraints.
58 
59### Documentation Requests
60 
61Bot requests for additional docs: Address if the code is genuinely unclear. Decline if
62the documentation would merely restate what the code already says clearly.
63 
64### Accessibility Improvements
65 
66Accessibility (ARIA labels, keyboard navigation, screen reader support) is a product
67priority decision that varies by project. Check project or user configuration for the
68team's stance. If no stance is declared, present the suggestion to the user with context
69about the scope and ask whether to address or decline.
70