Code Comment Quality and Dead Comment Review
Check that comments help rather than mislead.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Review comments in this code: [CODE_WITH_COMMENTS]. Classify comments as useful (why, not what), redundant, outdated, misleading, commented-out code, or TODOs without owners. Suggest which to delete, rewrite or turn into better names or tests. Check that public functions have clear descriptions of inputs, outputs and errors. Return a table with Location, Type, Problem and Fix, a count by type with percentages, and conventions for comments in the project. Add a worked count with invented comments: 40 comments; 14 explain why (35 percent); 12 repeat the code (30 percent); 6 are outdated (15 percent); 5 are commented-out code (12.5 percent); 3 are TODOs without owners (7.5 percent). Suggest deleting commented-out code since version control keeps history, turning repeat comments into better names and giving each TODO an owner and an issue link. Add a short convention: comments explain why, names explain what, tests explain behaviour, and TODOs carry an owner and a link to an issue. Add a lint rule idea for finding commented-out code and a pull request check for outdated comments near changed lines.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.
See an example result (fictional data)
Invented count of 40 comments. Explain why: 14 (35 percent). Repeat the code: 12 (30). Outdated: 6 (15). Commented-out code: 5 (12.5). TODO with no owner: 3 (7.5). Total 40. Actions: delete commented-out code (version control keeps history); rewrite repeating comments as better names; give each TODO an owner and an issue link. Convention: comments say why, names say what, tests say behaviour. Rule of thumb: if you need a comment to explain what a line does, try renaming first; keep the comment for why. Review after the next release.
Illustrative only, made-up data. Your results will differ and must be reviewed.