code-review

Automated code review for pull requests using multiple specialized agents

1.50x

Quality

—

Does it follow best practices?

Impact

99%

1.50x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Quality

Content

85%

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 highly actionable, well-sequenced review workflow with concrete commands, explicit validation/feedback loops, and a clean single-file structure. Its only real weakness is redundancy — the no-issues message and the false-positive guidance are each stated more than once.

Suggestions

Consolidate the 'No issues found' wording: state it once as a template and reference it from step 7 instead of repeating the text.

Merge the step-4 high-signal/Do-NOT-flag guidance with the trailing 'false positives, do NOT flag' list into a single checklist to remove the restatement.

Consider extracting the inline-comment URL formatting rules (currently several bullet lines) into a short named block referenced once, to trim the longest section.

Dimension	Reasoning	Score
Conciseness	The body is directive and avoids explaining concepts Claude already knows, but it repeats itself — the 'No issues found' comment text appears in both step 7 and the Notes template, and the high-signal/false-positive guidance is stated in step 4 then restated in the trailing list — so it is 'mostly efficient but could be tightened' rather than fully lean.	2 / 3
Actionability	It gives fully executable, copy-paste-ready guidance: concrete gh commands ('gh pr view <PR> --comments', 'gh pr review' with inline comments), specific agent tiers (haiku/sonnet/opus), and an exact inline-comment URL format with line-range rules, matching 'fully executable code/commands; specific examples'.	3 / 3
Workflow Clarity	The 9 steps are explicitly sequenced with validation checkpoints — a haiku gate in step 1, parallel issue validation in step 5, filtering of unvalidated issues in step 6 — providing the feedback loops and checklist the level-3 anchor requires.	3 / 3
Progressive Disclosure	No bundle files exist and none are needed; the single SKILL.md is well-organized into numbered steps and clearly delimited sections with easy navigation, which for a self-contained skill satisfies the level-3 'well-organized' bar.	3 / 3
	Total	11 / 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 names a clear domain and action, but it omits any 'Use when' trigger guidance and only offers a couple of natural keywords without variations. It is functional but generic enough to risk overlap with related skills.

Suggestions

Add an explicit trigger clause, e.g. 'Use when reviewing a GitHub pull request for bugs and CLAUDE.md compliance.'

Broaden natural trigger terms to include common variations users say, such as 'PR', 'review my pull request', or 'check this diff'.

List a couple more concrete actions (e.g. 'flags high-signal bugs and CLAUDE.md violations, posts inline review comments') to lift specificity from 'some actions' to 'multiple specific actions'.

Dimension	Reasoning	Score
Specificity	It names the domain and a concrete action — 'Automated code review for pull requests using multiple specialized agents' — but does not list multiple specific concrete actions, so it stops at the 'names domain and some actions' anchor rather than reaching 3.	2 / 3
Completeness	It answers 'what' (automated code review via specialized agents) but provides no 'Use when...' clause or equivalent explicit trigger guidance, which per the guidelines caps completeness at 2.	2 / 3
Trigger Term Quality	'code review' and 'pull requests' are natural terms a user would say, but coverage is thin with no common variations (e.g. 'PR', 'review my changes'), matching 'some relevant keywords but missing common variations' rather than the full-coverage level 3.	2 / 3
Distinctiveness Conflict Risk	'code review for pull requests' is a recognizable niche, but without explicit trigger terms it could still overlap with general coding/review skills, fitting 'somewhat specific but could still overlap' rather than the clearly-distinct level 3.	2 / 3
	Total	8 / 12 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
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	15 / 16 Passed

Repository: pipecat-ai/pipecat
Commit: 037ee5d

Reviewed: 4 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.