local-review-codex

Run the CI Codex PR review locally against this branch's unpushed work (committed + uncommitted) before pushing. Same policy and reasoning effort as the codex-pr-review GitHub action, on a newer model.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

82%Weight 40%Scale 1-5

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

The body is a lean, actionable, well-structured overview that defers implementation to run.sh and REVIEW.md while giving exact commands and a clear run sequence. Its only weak spots are minor: a little justifying prose and the lack of an explicit feedback loop and a formal reference-navigation block.

Suggestions

Trim the rationale sentences in 'Relaying the result' (e.g., 'the value of a cold Codex pass is surfacing what the current session would rationalize away') to lift conciseness toward a 5.

Add an explicit feedback loop for the version/auth precondition: e.g., 'If run.sh reports a CLI version error, upgrade with npm install --global @openai/codex@0.153.4 and re-run.'

Add a short 'References' block listing run.sh, REVIEW.md, and .github/workflows/codex-pr-review.yml with one-line purposes to make progressive disclosure navigation explicit.

Dimension	Reasoning	Score
Conciseness	The body is dense and information-rich with no padding about basic concepts, but a few rationale sentences (e.g., 'the value of a cold Codex pass is surfacing what the current session would rationalize away') could be trimmed; not a 5 because of these minor over-explanations.	4 / 5
Actionability	Provides copy-paste-ready commands (exact run.sh invocations, the precise upgrade command 'npm install --global @openai/codex@0.153.4', and the merge-base git flow) that cover the common cases fully.	5 / 5
Workflow Clarity	Clear sequence of Prerequisites -> Run -> Relaying the result, with a built-in version check as a validation checkpoint; not a 5 because there is no explicit feedback loop for handling a failed version check or auth issue beyond the upgrade hint.	4 / 5
Progressive Disclosure	Well-organized into clear sections with implementation logic deferred to run.sh and REVIEW.md as one-level-deep references; not a 5 because navigation to those files is mentioned inline rather than via a structured 'see X for details' map.	4 / 5
	Total	17 / 20 Passed

Description

78%Weight 40%Scale 1-5

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 specific, well-scoped description that clearly conveys both what the skill does and when to reach for it, with strong distinctiveness from sibling review skills. It sits just below top marks on specificity, trigger breadth, and completeness rather than exhibiting any real weakness.

Dimension	Reasoning	Score
Specificity	Names the concrete action ('Run the CI Codex PR review locally') with precise scope ('this branch's unpushed work (committed + uncommitted)') and timing ('before pushing'); not a 5 because it centers on a single action rather than multiple distinct capabilities.	4 / 5
Completeness	Explicitly states what it does (run the CI Codex PR review locally) and when to use it ('before pushing'); the 'when' is present and concrete but embedded in the action rather than a dedicated 'Use when...' clause, keeping it just below a 5.	4 / 5
Trigger Term Quality	Includes natural trigger terms users would say ('PR review', 'Codex', 'before pushing', 'CI'); good coverage but misses some common synonyms like 'pre-push' or 'lint my diff', so not a 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche ('CI Codex PR review, locally, on a newer model') and explicitly contrasts with the Claude-native 'local-review' skill, minimizing conflict risk.	5 / 5
	Total	17 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: windmill-labs/windmill
Path: .agents/skills/local-review-codex/SKILL.md
Commit: 7915540

Reviewed: 3 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.