Skip to content
Skills Library
Programming · Review & audit

reviewing-code

/reviewing-code

Reviews changes and incoming feedback for concrete defects, contract gaps, runtime cost, and useful simplification. Use for code review or after implementation; repository audits require an explicit request.

Repository
pragma ↗
Licence
MIT
Last repository change
26.10.2026

Assessment and security

90/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content25 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
3
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 26.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Reviews changes and incoming feedback for concrete defects, contract gaps, runtime cost, and useful simplification. Use for code review or after implementation; repository audits require an explicit request.

Original author description

Reviews changes and incoming feedback for concrete defects, contract gaps, runtime cost, and useful simplification. Use for code review or after implementation; repository audits require an explicit request.

How it works

  1. Find problems that matter
  2. Review the completed result
  3. Report
Original SKILL.md headings · 29 lines, 306 words
SKILL.md preview
---
name: reviewing-code
description: Reviews changes and incoming feedback for concrete defects, contract gaps, runtime cost, and useful simplification. Use for code review or after implementation; repository audits require an explicit request.
---

# Reviewing Code

Establish intended behavior, settled decisions, and review scope. For incoming feedback, start with specific claims and affected code. For a change review, use the named diff/files/PR; if unspecified, use pending changes or, when none, the last commit. Resolve material ambiguity before choosing the comparison range.

Read connected **production code, callers, and contracts before tests**. Trace consequential data/lifecycle paths and expected load. For change reviews, inspect the diff and limit the assessment to what it modifies or affects, including unchanged consumers.

## Find problems that matter

Assess correctness, completeness, safety/compatibility, runtime cost under expected load, and worthwhile simplification. Check meaningful coverage and brittle, duplicated, or implementation-coupled tests; use [tests-first](../tests-first/SKILL.md) for concrete evidence gaps.

Ground findings in reachable failures, violated contracts, or concrete runtime or maintenance costs. Confirm callers, inputs, usage, and scale. Thin adapters, single implementations, or documented shortcuts alone are not defects; broken contracts still are. Exclude stylistic preferences and pattern mandates without practical impact.

Read conditional guidance only for the relevant topic:

- [Architecture](../design-grill/references/architecture.md): material boundary/interface or restructuring concern.
- [Observability](references/observability.md): signal semantics, instrumentation, dashboards, or alerts.
- [Repository/debt audit](references/repository-audit.md): explicitly requested subsystem/repository audit or documented-deferral assessment.

## Review the completed result

Assess the whole change once after implementation: brief self-review for small, clear changes; independent review when risk or complexity justifies its cost. Individual delegated tasks may warrant earlier independent review. Reuse valid evidence while checking interactions and subsequent edits; re-review significant fixes and their effects as needed.

Verify feedback against code and constraints before acting. Clarify consequential uncertainty; retain evidence-backed disagreement. Suggestions do not override settled user choices or authorize out-of-scope fixes.

## Report

Return concise, prioritized, deduplicated findings with location, practical consequence, and the smallest useful remedy. Separate confirmed defects from consequential uncertainty and state coverage limits. Report no findings honestly; review alone authorizes neither edits, publication, nor shipping/merging.
Files in this package (3)
  • SKILL.md 3 KB
  • references/observability.md 2 KB
  • references/repository-audit.md 2 KB

Versions

RepositoryScoreLicence
This page · bbeny123/pragma90MIT
joeblackwaslike/agent-skills100MIT
Montinou/triqual100MIT
thesoftwarehouse/agentic-collections100MIT
alexei-led/cc-thingz100MIT
julianobarbosa/claude-code-skills100MIT
isvlasov/rageatc-oss100MIT
bostonaholic/rpikit100MIT

Database snapshot: 13.11.2026 · package version 901dcf282966 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 90code-review
    Use when reviewing a bounded diff or pull request for actionable defects and regressions.
  • 90doompi-review
    Review a code change for concrete defects, regressions, missing tests, and contract violations. Use when the user asks for code review, diff review, or risk assessment, not as a general implementation workflow.
  • 100code-review
    Reviews code changes for bugs, style issues, security vulnerabilities, and best practices. Use when reviewing PRs, checking code quality, or validating implementations.
  • 90reviews
    Review a pull request or code change for code health. Use when a user asks to "code review", "review this PR", or "review this change".
  • 100code-review
    Review C code changes for correctness, safety, and style. Use this when you want to review C code changes or PRs.
  • 90code-review
    Review Proto Fleet pull requests for concrete security, correctness, and reliability defects. Use for every GitHub Copilot code review.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.