ce-code-review

Structured code review for bugs, regressions, tests, and standards. Use before PRs or when asked for review; report-only by default, with explicit local apply available for user-directed fix workflows.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

85%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-architected complex orchestration skill: clear staged workflow with validation gates, executable commands, and clean one-level-deep progressive disclosure. The main weakness is conciseness — the prose is long and over-qualified where tighter phrasing would preserve all the operational precision.

Suggestions

Tighten the longest sentences in Stage 3d and the execution-spine steps by splitting compound clauses and dropping redundant 'never/must' restatements that are already implied by earlier rules.

Consolidate the repeated scope-mode prohibition prose (pr-remote/branch-remote 'must not Read/Grep workspace paths') into one stated rule referenced from each path instead of restating it three times.

Move the dense git/scope-detection command blocks into a reference file and keep only the decision logic inline, mirroring the on-demand pattern already used for persona and dispatch references.

Dimension	Reasoning	Score
Conciseness	The body avoids concept-padding and is operationally dense, but at ~556 lines it carries many long, heavily qualified sentences and repeated "never/must" caveats that could be tightened without losing meaning, placing it at 'mostly efficient but could be tightened' rather than 'every token earns its place'.	2 / 3
Actionability	Provides copy-paste-ready, executable bash (gh pr view, git merge-base, git diff -U10, the review-scope.py invocation, scratch-root setup) plus explicit argument, output, and severity tables — concrete and specific rather than abstract.	3 / 3
Workflow Clarity	A clearly ordered staged spine (1 through 6) with explicit fail-closed validation gates (Stage 1b lite eligibility, Stage 3c, Stage 3d 'do not proceed until roster materialized') and feedback loops (peer-job recovery, re-validate), satisfying the checkpoint requirement for batch/apply operations.	3 / 3
Progressive Disclosure	The body is an overview that loads detailed references on demand; the References table maps each one-level-deep file to its load stage and purpose, and every referenced file (persona-catalog.md, dispatch-reviewers.md, finish-review.md, etc., plus personas/) exists in the bundle.	3 / 3
	Total	11 / 12 Passed

Description

90%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: third-person voice, explicit what-and-when, and natural trigger terms. The only soft spot is specificity, which enumerates review targets and modes rather than a breadth of distinct concrete actions.

Dimension	Reasoning	Score
Specificity	Names the domain and review targets ("bugs, regressions, tests, and standards") plus report/apply modes, but these are review targets rather than a list of multiple distinct concrete actions, so it sits at 'names domain and some actions' rather than the top anchor.	2 / 3
Completeness	Explicitly answers both what ("Structured code review ... report-only by default, with explicit local apply available") and when ("Use before PRs or when asked for review") with an explicit trigger clause.	3 / 3
Trigger Term Quality	Includes natural terms a user would actually say — "code review", "bugs, regressions, tests", "PRs", "when asked for review" — giving good coverage of common phrasings.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (structured pre-PR code review) with distinct triggers ("before PRs", "when asked for review") and a report-only default that separates it from generic apply-style skills.	3 / 3
	Total	11 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (556 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: EveryInc/compound-engineering-plugin
Path: skills/ce-code-review/SKILL.md
Commit: a9f6d53

Reviewed: 7 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.