Name: autoreview
Rating: 56.8 (1 reviews)
Author: openclaw

Blog Docs Log in Get started

autoreview

Pre-commit/ship code review: Codex default; optional Claude or Pi.

Quality

63%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/autoreview/SKILL.md

The canonical home for this skill is autoreview in openclaw/openclaw

Quality

Content

77%

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

The body is highly actionable with executable commands and clear validated workflows, but it is verbose and monolithic, keeping large reference-worthy material inline in SKILL.md rather than progressively disclosing it into separate files.

Suggestions

Move the engine model/thinking tables and engine-isolation flag tables into a separate reference file (e.g. ENGINES.md) and link to it from SKILL.md to reduce inline bulk.

Tighten the Contract and Scope Governor prose by collapsing repetitive scope-break bullets into a compact checklist.

Deduplicate model default information that appears both in the intro paragraph and again in the 'Models and thinking' section.

Dimension	Reasoning	Score
Conciseness	The body is information-dense with domain-specific operational policy rather than concepts Claude already knows, but it is very long with verbose prose sections (Contract, Scope Governor, Release Branches) and some repeated model/engine detail that could be tightened, so it is mostly efficient rather than lean.	2 / 3
Actionability	It provides fully executable, copy-paste-ready commands ('$AUTOREVIEW --mode branch --base origin/main'), concrete env-var tables, and platform-specific path-setup blocks for both bash and PowerShell, with specific flags and examples throughout.	3 / 3
Workflow Clarity	Multi-step processes are explicitly sequenced with validation checkpoints and feedback loops, e.g. 'rerun focused tests and rerun the structured review helper', the two-cycle scope pause with reclassification, and 'stop as soon as the helper exits 0 with no accepted/actionable findings'.	3 / 3
Progressive Disclosure	Section headers and the real bundled scripts ($AUTOREVIEW, $AUTOREVIEW_HARNESS, which exist in scripts/) provide structure and navigation, but the skill is a single large monolithic SKILL.md with model tables, engine-isolation detail, and policy prose all inline rather than split into one-level-deep reference files.	2 / 3
	Total	10 / 12 Passed

Description

50%

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

The description is concise and third-person, naming a clear action and engine defaults, but it omits explicit 'Use when' trigger guidance and only partial natural-language trigger coverage, leaving every dimension at the midpoint.

Suggestions

Add an explicit 'Use when...' clause naming natural triggers like 'when the user asks for a Codex/Claude/Pi review or a second-model review before commit or ship'.

Expand trigger terms in the description to include common user phrasings ('PR review', 'review before commit', 'second opinion') instead of relying on the body for them.

Tighten distinctiveness by contrasting with sibling skills inline (e.g. 'code review closeout, not Guardian auto_review approval routing').

Dimension	Reasoning	Score
Specificity	The phrase 'Pre-commit/ship code review' names a concrete domain and action, plus the engine options 'Codex default; optional Claude or Pi', but it lists a single action rather than multiple specific concrete actions, so it does not reach level 3.	2 / 3
Completeness	It clearly states what the skill does ('Pre-commit/ship code review' with engine defaults) but lacks any 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 2 per the rubric.	2 / 3
Trigger Term Quality	'code review' and the engine names 'Codex/Claude/Pi' are natural terms a user might say, but common variations like 'review before commit', 'second-model review', or 'PR review' appear only in the body, not the description, so coverage is partial.	2 / 3
Distinctiveness Conflict Risk	The pre-commit/ship framing plus named engines carve out a somewhat specific niche, but 'code review' alone is broad enough to overlap with other review-oriented skills, so it is not yet a clearly conflict-free niche.	2 / 3
	Total	8 / 12 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: openclaw/crabbox
Path: .agents/skills/autoreview/SKILL.md
Commit: f7c9155

Reviewed: about 20 hours 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.