@@ -7,6 +7,59 @@ pull request review. Path-specific guidance lives under `.github/instructions/`.
77ALWAYS acknowledge in the review summary that these provider review
88instructions are being used.
99
10+ ## Severity and Nit Policy (read first)
11+
12+ This repository is ** community-maintained** . Contributor friction is the
13+ single biggest cost to the project. Review feedback must respect that.
14+
15+ ### Only report HIGH and MEDIUM findings
16+
17+ - Report ` HIGH ` : correctness bugs, regressions, breaking schema/state
18+ changes without migration, security issues, secret leakage, panics,
19+ data loss risks.
20+ - Report ` MEDIUM ` : missing test coverage for changed behavior, missing
21+ example for a new resource, missing docs update for a schema change,
22+ missing ` Sensitive: true ` on secret-bearing attributes, missing
23+ ` Description ` on schema attributes, missing ` ValidateFunc ` /
24+ ` ValidateDiagFunc ` on bounded inputs, missing import docs.
25+ - ** Do not report ` LOW ` findings or nits.** If the only thing you would
26+ say is ` LOW ` , say nothing.
27+
28+ ### Do NOT comment on (defer to linters / human reviewers)
29+
30+ - Code formatting, whitespace, import ordering, line length.
31+ - Naming preferences, identifier style, comment wording, doc prose
32+ polish, grammar.
33+ - "Consider extracting…", "this could be a helper", or other speculative
34+ refactors that are not requested by the change.
35+ - Style of existing surrounding code the PR did not touch.
36+ - Adding comments, docstrings, or type hints to code the PR did not
37+ change.
38+ - Test naming conventions or alternative test framings when the
39+ existing test adequately covers the behavior.
40+ - Hypothetical errors that cannot occur given the call sites.
41+
42+ ### Always report even if it looks like a nit
43+
44+ These items affect end-user Terraform behavior and must be flagged as at
45+ least ` MEDIUM ` regardless of how small they look:
46+
47+ - Secret-bearing attribute missing ` Sensitive: true ` .
48+ - Schema attribute missing ` Description ` .
49+ - Bounded input missing ` ValidateFunc ` /` ValidateDiagFunc ` .
50+ - New resource/data source without at least one example under
51+ ` examples/ ` or docs under ` website/ ` .
52+ - Behavior change without a corresponding test change.
53+ - Resource that supports import but has no documented import ID format.
54+
55+ ### Output discipline
56+
57+ - If there are no HIGH or MEDIUM findings, the review must say
58+ ` No blocking findings found ` and stop. Do not pad with low-value
59+ observations.
60+ - Keep each finding to its impact, file reference, and a concise fix.
61+ Do not lecture, restate the diff, or suggest unrelated improvements.
62+
1063## Review Goals
1164
1265- Find correctness bugs, regressions, and provider behavior changes.
@@ -143,18 +196,20 @@ Use this background when judging schema, examples, or state changes.
143196
144197## Review Report Format
145198
146- Return findings first, ordered by severity:
199+ Return findings first, HIGH before MEDIUM (no LOW — see Severity and Nit
200+ Policy above):
147201
148- 1 . ` HIGH ` /` MEDIUM ` / ` LOW ` title — short impact statement
202+ 1 . ` HIGH ` /` MEDIUM ` title — short impact statement
1492032 . File reference: ` path/to/file.go:line `
1502043 . Why this is a problem (runtime behavior, Terraform UX, upgrade risk)
1512054 . Suggested fix (concise)
152206
153207Then include:
154208
155- - ` Open Questions / Assumptions `
156- - ` Residual Risk `
209+ - ` Open Questions / Assumptions ` (only if non-trivial)
210+ - ` Residual Risk ` (only if non-trivial)
157211- ` Change Summary ` (brief)
158212
159- If no issues are found, explicitly state ` No blocking findings found ` and list
160- remaining risk areas.
213+ If no HIGH or MEDIUM findings exist, state ` No blocking findings found `
214+ and stop. Do not add nit sections, style observations, or speculative
215+ suggestions.
0 commit comments