code-review-bot

Use this skill when you want to automatically review GitHub pull requests and post line-level comments with findings, suggestions, and approval recommendations.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./external/goose-recipes/code-review-bot/SKILL.md

Quality

Content

72%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 highly actionable with complete executable code and a useful reference table, but it lacks validation/verification before posting to a PR and keeps a large script monolithically inline with no progressive disclosure to reference files.

Suggestions

Add a validation/dry-run checkpoint before post_review (e.g. print findings for confirmation, or guard create_review behind an explicit approve flag) so the outward-facing posting step is not unvalidated.

Move the review engine into a scripts/ file and reference it from SKILL.md to improve progressive disclosure and reduce inline bulk.

Replace the hard-coded 'claude-opus-4-6' model name with a current model or a configurable variable to avoid stale version-pin drift.

Dimension	Reasoning	Score
Conciseness	Mostly efficient: a brief Overview, minimal setup, executable code, and a compact Quick Reference table with no padding or explanations of concepts Claude already knows; held at 4 rather than 5 because the full inline engine script and a hard-coded model name ('claude-opus-4-6') could be tightened or factored out.	4 / 5
Actionability	Fully executable, copy-paste-ready Python with concrete imports, a complete tool schema, and a Quick Reference table covering each finding category — concrete guidance that covers the common cases.	5 / 5
Workflow Clarity	The fetch-diff → review-each-file → post-review sequence is clear, but posting review comments (and a REQUEST_CHANGES event) to a PR is an outward-facing batch operation with no validation, dry-run, or confirmation checkpoint before create_review, which caps the score at 3 per the destructive/batch guideline.	3 / 5
Progressive Disclosure	Sections are present (Overview, Setup, PR Review Engine, Quick Reference) with no nested or buried references, but the entire review engine is one large inline script and no bundle/reference files exist, so content that could be split out is inlined.	3 / 5
	Total	15 / 20 Passed

Description

87%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, complete, and clearly distinct, with explicit what-and-when guidance. The main weakness is second-person voice and slightly incomplete trigger-term synonym coverage.

Suggestions

Rewrite in third person (e.g. 'Automatically reviews GitHub pull requests...') to avoid the second-person voice penalty.

Add common synonyms users would say — 'PR', 'code review', 'diff' — to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — 'review GitHub pull requests', 'post line-level comments with findings, suggestions, and approval recommendations' — giving comprehensive coverage; reduced from 5 to 4 because the second-person phrasing 'Use this skill when you want to' triggers the voice penalty.	4 / 5
Completeness	Explicitly answers both what (review PRs, post line-level comments with findings/suggestions/approval recommendations) and when ('Use this skill when you want to automatically review GitHub pull requests') with a concrete trigger phrase.	5 / 5
Trigger Term Quality	Good natural keywords ('GitHub pull requests', 'review', 'line-level comments', 'findings', 'suggestions') but missing common synonyms users would say like 'PR', 'code review', or 'diff', so it falls short of the comprehensive anchor.	4 / 5
Distinctiveness Conflict Risk	Targets a clear niche — automated GitHub PR review with line-level comment posting — with distinct triggers and minimal overlap risk with unrelated skills.	5 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: ProwlrBot/prowlr-marketplace
Path: external/goose-recipes/code-review-bot/SKILL.md
Commit: 32f0050

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.