autoreview

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

Quality

65%

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

Fix and improve this skill with Tessl

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

Quality

Content

70%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 content is highly actionable with strong, explicit feedback loops and concrete commands, but it is noticeably verbose with time-sensitive version strings inline and inlines reference-grade material that would benefit from separate files. Progressive disclosure is the weakest dimension.

Suggestions

Move the 'Review engine isolation' prose and the engine model/thinking tables into a dedicated reference file (e.g. references/engines.md) and link to it one level deep to improve progressive disclosure.

Collect version requirements (v2.1.169+, v0.79.0+, deprecated package names) into a single 'Compatibility / deprecated' section so time-sensitive details stop penalizing conciseness throughout the body.

Tighten the densest policy paragraphs (Scope Governor, engine isolation) into shorter bullet statements to reduce token load without losing the validation guidance.

Dimension	Reasoning	Score
Conciseness	The body is a ~345-line reference manual with noticeably padded sections (the multi-paragraph 'Review engine isolation' wall) and many inline time-sensitive version numbers (v2.1.169+, v0.79.0+, gpt-5.5) not placed in a deprecated/old-patterns section, which the rubric penalizes; it does not explain basic concepts, but it is verbose rather than lean.	2 / 5
Actionability	Commands are copy-paste ready across the common cases (local, branch, commit, panels, per-engine model/thinking flags) with real flags and a dedicated path-setup section resolving the $AUTOREVIEW substitution; examples cover the common cases fully.	5 / 5
Workflow Clarity	Explicit validation steps and feedback loops are present throughout ("rerun focused tests and rerun the structured review helper", "Stop as soon as the helper exits 0 with no accepted/actionable findings", scope-break pause after two non-converging cycles), with error-recovery paths for malformed caches and capacity retries.	5 / 5
Progressive Disclosure	The bundle is scripts-only (autoreview, test-review-harness) and those paths are signaled via the 'Skill Path' section, but the bulk of reference material — the giant isolation-flags paragraph and the engine model/thinking tables — is inlined in SKILL.md rather than split into one-level-deep reference files, leaving structure present but not well separated.	3 / 5
	Total	15 / 20 Passed

Description

61%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.

The description is concrete and well-scoped with good engine keyword coverage, but lacks an explicit 'Use when...' trigger clause, which caps completeness and keeps it from a top score. It is distinct and actionable but slightly terse on the 'when'.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when the user asks for a code review or second-model review before commit, push, or ship.'

Expand the action coverage beyond the single 'code review' phrase (e.g. 'runs a structured review and validates findings') to lift specificity.

Include a couple of natural synonyms users say, such as 'PR review' or 'second opinion', to round out trigger terms.

Dimension	Reasoning	Score
Specificity	"Pre-commit/ship code review" names the domain and one concrete action; the enumerated engines (Codex, Claude, Pi, Droid, Copilot, OpenCode) are concrete nouns but not distinct actions, so it stops at 'names domain and 1-2 concrete actions, not comprehensive' rather than 'lists several specific actions'.	3 / 5
Completeness	The 'what' is clear ("Pre-commit/ship code review") but the 'when' is only weakly implied by "Pre-commit/ship" with no explicit 'Use when...' clause, which per the rubric caps completeness at 3.	3 / 5
Trigger Term Quality	Natural terms like "code review", "Codex review", "Claude review", and the other named engines give good keyword coverage a user would actually say; a few common synonyms (e.g. "PR review", "second opinion") are missing, so it sits just below comprehensive.	4 / 5
Distinctiveness Conflict Risk	The engine-specific closeout framing ("Codex default; optional Claude, Pi, Droid, Copilot, or OpenCode") gives it a mostly-distinct niche with only minor overlap risk against generic code-review skills, not the clear minimal-conflict niche of a 5.	4 / 5
	Total	14 / 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: udecode/plate
Path: .agents/skills/autoreview/SKILL.md
Commit: cce36d3

Reviewed: 6 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.