/basic-review
A portable correctness review. It reads the diff and reports; it never edits,
stages or commits. A caller that fixes findings (/qa-full Step 3) applies the
fixes itself and re-runs this review to verify them.
The bar for what counts as a blocker is the superskills quality standard,
ENGINEERING_STANDARDS.md in the superskills install (a plugin install exposes
it at ${CLAUDE_PLUGIN_ROOT}/ENGINEERING_STANDARDS.md). It is not expected in
the project under review.
Scope
- With an argument, review exactly that: a range such as
origin/main...HEAD
(as /qa-full passes it), or paths.
- Without one, "pending changes" means the current branch's diff against its
base plus anything uncommitted:
git diff <base>...HEAD and git diff HEAD.
Find the base the way /qa-full Step 1 does (gh repo view default branch,
else origin/HEAD, else main/master).
- If there is nothing to review, say so and stop.
Read each changed hunk with enough surrounding code (callers, the function it
sits in, the config it reads) to judge it. Verify claims against the actual
code rather than inferring from names.
Review
Review the pending changes in this repo. Flag issues a careful maintainer would
block on before merge, in roughly this order:
- Correctness: logic errors, broken invariants, missed edge cases,
off-by-one, races, lifetime/memory bugs, error paths that swallow failures.
- Security: injection, authn/authz mistakes, secret handling, unsafe
deserialization, unsafe defaults, TOCTOU.
- Reliability & performance: blocking the wrong thread, unbounded resource
use, accidental O(N²), retries without backoff, silent failure paths.
- Contract risk: behavior changes callers rely on, silently changed
defaults, broken backward compatibility, missing migrations.
Then apply checklist.md in this skill's directory (gstack's pre-landing
checklist): run its Pass 1 (CRITICAL) categories, then Pass 2, and honor its
Suppressions list. This skill is read-only, so ignore the checklist's
Fix-First / AUTO-FIXED instructions and its output format (use the Report format
below), and skip the items it assigns to gstack's specialist subagents or to
gstack-only markers.
For each finding cite file:line, explain why it is wrong (not what the code
does), and propose a concrete fix when one is obvious. Be calibrated: if a
finding is not high-confidence, say so or skip it.
What earns a finding
A candidate earns a finding only if every one of these holds:
- It meaningfully affects correctness, security, performance, or
maintainability, and it is discrete and actionable — one defect, one fix.
- The diff introduced it. Problems that predate the change get at most a
one-line aside labelled "pre-existing" — never a finding, never a blocker.
- The rigor demanded matches the file's neighborhood: a corner of one-off
scripts does not need enterprise-grade validation.
- The author of this change would fix it if told; if they would reasonably
argue, it is not a finding.
- No speculation: a harm of the form "may break other callers" needs the
caller that is provably affected, named — or it is dropped.
- It is not an intentional behavior change dressed up as a bug.
If nothing clears this bar, "No findings worth fixing in " is the
correct and complete review. An empty result beats an invented one.
Skip pure style, formatting, naming, and anything a linter or CI already
enforces.
Report
One entry per finding, most severe first:
[CRITICAL|HIGH|MEDIUM|LOW] (confidence: high|medium) file:line
Why it is wrong: ...
Fix: ... (or "no obvious fix")
CRITICAL and HIGH are what a caller treats as blockers. End with one line naming
the range reviewed and the number of findings by severity.
If you find nothing worth fixing, say so plainly: "No findings worth fixing in
."
Related
/review (gstack): the full review. Prefer it when installed; it goes deeper
and applies fixes itself.
/clean-code: KISS/DRY/SOLID/YAGNI quality cleanups, which this review skips.
/defense: the deeper security audit of the same diff.
1---2name: basic-review3version: 1.2.04description: |5 Read-only correctness review of pending changes, needing no external6 install: correctness, security, reliability/performance and contract risk,7 each finding cited as file:line with why it is wrong and a fix. The fallback8 when gstack's /review is unavailable (not installed, or its install is9 broken); /qa-full and /daily-qa call it automatically then. Applies a10 vendored copy of gstack's pre-landing checklist. Use when asked for11 a "basic review", a "quick review without gstack", or a review in a tool where12 /review is missing. Not for quality cleanups (use /clean-code), and not the13 first choice when /review is installed: /review goes deeper and applies fixes.14allowed-tools:15 - Bash16 - Read17 - Grep18 - Glob19metadata:20 source: https://skillmd.com/skills/ariadoss/basic-review21---2223# /basic-review2425A portable correctness review. It reads the diff and reports; it never edits,26stages or commits. A caller that fixes findings (`/qa-full` Step 3) applies the27fixes itself and re-runs this review to verify them.2829The bar for what counts as a blocker is the superskills quality standard,30`ENGINEERING_STANDARDS.md` in the superskills install (a plugin install exposes31it at `${CLAUDE_PLUGIN_ROOT}/ENGINEERING_STANDARDS.md`). It is not expected in32the project under review.3334## Scope3536- With an argument, review exactly that: a range such as `origin/main...HEAD`37 (as `/qa-full` passes it), or paths.38- Without one, "pending changes" means the current branch's diff against its39 base plus anything uncommitted: `git diff <base>...HEAD` and `git diff HEAD`.40 Find the base the way `/qa-full` Step 1 does (`gh repo view` default branch,41 else `origin/HEAD`, else `main`/`master`).42- If there is nothing to review, say so and stop.4344Read each changed hunk with enough surrounding code (callers, the function it45sits in, the config it reads) to judge it. Verify claims against the actual46code rather than inferring from names.4748## Review4950Review the pending changes in this repo. Flag issues a careful maintainer would51block on before merge, in roughly this order:52531. **Correctness:** logic errors, broken invariants, missed edge cases,54 off-by-one, races, lifetime/memory bugs, error paths that swallow failures.552. **Security:** injection, authn/authz mistakes, secret handling, unsafe56 deserialization, unsafe defaults, TOCTOU.573. **Reliability & performance:** blocking the wrong thread, unbounded resource58 use, accidental O(N²), retries without backoff, silent failure paths.594. **Contract risk:** behavior changes callers rely on, silently changed60 defaults, broken backward compatibility, missing migrations.6162Then apply `checklist.md` in this skill's directory (gstack's pre-landing63checklist): run its Pass 1 (CRITICAL) categories, then Pass 2, and honor its64**Suppressions** list. This skill is read-only, so ignore the checklist's65Fix-First / AUTO-FIXED instructions and its output format (use the Report format66below), and skip the items it assigns to gstack's specialist subagents or to67gstack-only markers.6869For each finding cite `file:line`, explain *why* it is wrong (not what the code70does), and propose a concrete fix when one is obvious. Be calibrated: if a71finding is not high-confidence, say so or skip it.7273### What earns a finding7475A candidate earns a finding only if every one of these holds:7677- It meaningfully affects correctness, security, performance, or78 maintainability, and it is discrete and actionable — one defect, one fix.79- **The diff introduced it.** Problems that predate the change get at most a80 one-line aside labelled "pre-existing" — never a finding, never a blocker.81- The rigor demanded matches the file's neighborhood: a corner of one-off82 scripts does not need enterprise-grade validation.83- The author of this change would fix it if told; if they would reasonably84 argue, it is not a finding.85- No speculation: a harm of the form "may break other callers" needs the86 caller that is provably affected, named — or it is dropped.87- It is not an intentional behavior change dressed up as a bug.8889If nothing clears this bar, "No findings worth fixing in <range>" is the90correct and complete review. An empty result beats an invented one.9192Skip pure style, formatting, naming, and anything a linter or CI already93enforces.9495## Report9697One entry per finding, most severe first:9899```100[CRITICAL|HIGH|MEDIUM|LOW] (confidence: high|medium) file:line101Why it is wrong: ...102Fix: ... (or "no obvious fix")103```104105CRITICAL and HIGH are what a caller treats as blockers. End with one line naming106the range reviewed and the number of findings by severity.107108If you find nothing worth fixing, say so plainly: "No findings worth fixing in109<range>."110111## Related112113- `/review` (gstack): the full review. Prefer it when installed; it goes deeper114 and applies fixes itself.115- `/clean-code`: KISS/DRY/SOLID/YAGNI quality cleanups, which this review skips.116- `/defense`: the deeper security audit of the same diff.