Read the comment as a claim#
A comment that says "this may cause an N+1 query" asserts two things: something per row reads a related object, and the query feeding the rows does not load it in advance. A comment that says "this value might be None" asserts that some caller, default or data source can produce the None. Neither assertion is settled by how confident the comment sounds, by its severity label or by how much code it quotes. Each is settled by a fact you can usually establish in minutes: a query count, a declared type, a timing, a test run at a boundary.
Three outcomes follow from that fact. When it confirms the claim, change the code. When reading cannot settle it but a run can, write the test first, because a test that fails before the fix and passes after it answers the comment and keeps the answer. When the fact disproves the claim, dismiss the comment and put the fact in your reply, so the next reader does not have to find it again.
One pull request, two answers#
A September 2026 pull request in Gyrinx, a Django application, shows both ends. It drew each user's badge beside their name on campaign screens, and Copilot reviewed it several times. Two of its comments made the same kind of claim.
One comment said a shared row template was also rendered by the signed-in dashboard, whose query did not load the owner's badges, "so a page with several campaigns regresses to N+1." Copilot review The author counted: "the dashboard costs 20 queries with one campaign listed and 20 with five (warm session, CaptureQueriesContext), so nothing grows with the number of rows." Author's reply A flat count disproves an N+1 claim, so the comment was declined with the numbers.
The other said a refused form submission redrew the page and loaded the badge data lazily. The author confirmed it and wrote a test that compares the redraw's query count with the GET's. It "fails against the reviewed guard (23 vs 21) and passes with the join." Author's reply Same reviewer, same kind of claim, same deciding fact; one comment earned a dismissal and the other a change with its test. The N+1 guide follows both threads through the merged code.
Find the fact for each kind of claim#
The guides below each take one kind of comment, follow public review threads to their resolution and say which evidence points to a change, a test or a dismissal.
Correctness#
| Comment says | Fact that settles it | Guide |
|---|---|---|
| This might be null | What produces the value: its type, constructor and defaults | Null checks |
| Off by one | A test at the first value, the last value and one past it | Off-by-one |
| Race condition | A point where a second operation can run between two steps | Race conditions |
| Missing await | What the call returns and whether anything awaits it | Missing await |
| Overflow or truncation | The type's width and the largest value the code really sees | Integer overflow |
| Timezone bug | Whether the value is an instant or a wall-clock time | Timezones |
Errors and resources#
| Comment says | Fact that settles it | Guide |
|---|---|---|
| The error is swallowed | Whether anyone downstream needs to know it failed | Swallowed errors |
| The error is not handled | A realistic cause and the nearest layer that can act on it | Unhandled errors |
| This leaks | Who owns the resource and how long the owner lives | Resource leaks |
| Retrying is unsafe | What the server does when the same request arrives twice | Retries |
| Validate this input | Whether the value crosses a trust boundary here | Input validation |
Security#
| Comment says | Fact that settles it | Guide |
|---|---|---|
| SQL injection | Whether untrusted text reaches the SQL before it is parsed | SQL injection |
| Hardcoded secret | What the string lets someone who copies it do | Hardcoded secrets |
| Path traversal | Where the final path lands after every decoding step | Path traversal |
| The log leaks a token | What the logged value contains at runtime | Sensitive logging |
| ReDoS | Who controls the input, and timings in your regex engine | Regex ReDoS |
| XSS | Where the string comes from and which context it lands in | XSS |
Performance and data volume#
| Comment says | Fact that settles it | Guide |
|---|---|---|
| N+1 query | The query count at one row and at several | N+1 queries |
| Use a faster construct | How often the line runs and how large its input gets | Micro-optimizations |
| The cache can go stale | A write path that changes an input and skips invalidation | Cache invalidation |
| No limit on this query | How many rows it reads and how fast they grow | Unbounded queries |
| This grows without limit | The code that removes entries, and the bound it implies | Memory growth |
Compatibility and dependencies#
| Comment says | Fact that settles it | Guide |
|---|---|---|
| This breaks callers | Who the callers are and what you promised them | Breaking changes |
| The default changed | Who relies on the old default | Changed defaults |
| This API is deprecated | The source or changelog of the version you pin | Deprecations |
| This version is wrong | The registry, the changelog and your support policy | Dependency versions |
| Use this API instead | Whether it exists and behaves the same on your oldest supported version | Suggested APIs |
Suggestions and style#
| Comment says | Fact that settles it | Guide |
|---|---|---|
| A committable suggestion | Whether the replaced lines are complete on their own | Suggested changes |
| A suggested fix | What the code does on the named input before and after | Suggestions that add bugs |
| Rename this | Whether the name is false or breaks a written convention | Naming |
| Extract the duplicate | Whether the copies change for the same reason | Duplicate code |
| The docstring is stale | Which side the callers and tests rely on | Stale docstrings |
| Use the idiom | The language specification for the version you build with | Language idioms |
Tests#
| Comment says | Fact that settles it | Guide |
|---|---|---|
| Add a test | Whether a test already fails on the old code | Missing tests |
| This is a bug | A test with the claimed input, run before and after the fix | Regression tests |
| The assertion is weak | Whether the test fails when you put the defect in | Weak assertions |
| This test is flaky | The source of nondeterminism and whether varying it fails the test | Flaky tests |
| This mocks too much | Whether breaking the unit turns the test red | Over-mocked tests |
Check where the comment sits#
Some comments are decided less by their claim than by their place in the pull request. A comment on code the pull request did not change may describe a real problem that belongs in its own issue (comments on unchanged code). A comment GitHub marks outdated may or may not still apply to the current head (outdated comments). A question asks for an answer that becomes a change, a test or a recorded reason (question comments). Two reviewers can ask for opposite changes (conflicting comments), and a reviewer can flag the very behavior the pull request means to change (comments against the intended change).
Leave the fact in the thread#
Whichever way you decide, write the deciding fact where the comment is. A change links its commit and, where one exists, the test that fails without it. A dismissal names the type, the count, the caller or the version that disproves the claim, and says what you did not check. The dismissal reply guide shows how to write one that a later reader can verify. Teams scoring a reviewer across a trial can then count the thread as the dismissed finding guide describes, without asking the author again.