Skill

ctx-code-review

From ctx

Reviews code changes, diffs, or PRs for correctness, edge cases, naming clarity, test gaps, and conventions from .context/CONVENTIONS.md.

Git

code-quality

npx claudepluginhub activememory/ctx --plugin ctx

Tool Access

This skill is limited to using the following tools:

ReadGrepGlobBash(git:*)

Preview

Review the specified code change focusing on substance over style.

SKILL.md

Similar Skills

code-review

Automates code reviews on git diffs, staged changes, PRs, files, or branches against CLAUDE.md conventions and prioritized best practices (security first).

10 files

claude-skills

pr-review

221

Performs diff-based PR reviews across code quality, test coverage, silent failures, type design, and comment quality with severity-ranked findings on git diffs or specified PRs.

7 files

armory

code-reviewer

Reviews pull requests and code changes for correctness, security, performance, code quality, testing, documentation, and maintainability using git diffs.

5 files6 tools

charon-fan-agent-playbook

Stats

Parent Repo Stars41

Parent Repo Forks9

Last CommitMar 28, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

## Review: <scope> ### Issues - **[severity]** file:line - description ### Observations - Note anything non-obvious but not necessarily wrong ### Verdict One sentence: ship it, fix N issues first, or needs rethink.

Review the specified code change focusing on substance over style.

When to Use

User says "review this code", "review this change", "code review"
User asks for feedback on a diff, PR, or set of changes
User says "what do you think of this?"

When NOT to Use

User wants a full PR review with GitHub integration (use /gitnexus-pr-review instead)
User wants an architecture-level review (use /ctx-architecture)

Review Checklist

Work through each dimension. Flag issues, don't fix them unless asked.

Correctness: Does the logic do what it claims? Off-by-one errors, nil dereferences, race conditions?
Edge cases: What happens with empty input, max values, concurrent access, or partial failures?
Naming clarity: Do function, variable, and type names communicate intent without needing comments?
Test coverage gaps: What behavior is untested? What inputs would exercise uncovered paths?
Convention adherence: Does this follow the project patterns documented in .context/CONVENTIONS.md?

Execution

Read .context/CONVENTIONS.md to load project patterns
Identify the scope: file(s), diff, or recent changes
If no specific target, check git diff for unstaged changes
Work through each checklist dimension
Present findings grouped by severity: bugs > logic gaps > conventions > style observations

Output Format

Lead with the most important finding. Use this structure:

## Review: <scope>

### Issues
- **[severity]** file:line - description

### Observations
- Note anything non-obvious but not necessarily wrong

### Verdict
One sentence: ship it, fix N issues first, or needs rethink.

Flag but don't fix style issues. Focus review on substance over formatting.