github-code-review
Comprehensive GitHub code review with AI-powered swarm coordination
45
22%
Does it follow best practices?
Impact
85%
3.69xAverage score across 3 eval scenarios
Critical
Do not install without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/github-code-review/SKILL.mdQuality
Discovery
22%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description relies heavily on buzzwords ('comprehensive', 'AI-powered', 'swarm coordination') without specifying concrete actions or when the skill should be used. It mentions the domain (GitHub code review) but fails to enumerate specific capabilities or provide trigger guidance, making it difficult for Claude to reliably select this skill over alternatives.
Suggestions
Replace vague language with specific actions, e.g., 'Reviews pull request diffs, posts inline comments, checks for code style violations, and suggests improvements.'
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the user asks to review a PR, check a pull request, analyze a GitHub diff, or provide code review feedback.'
Remove or clarify buzzwords like 'swarm coordination' — either explain what it means concretely or replace with actionable language.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses vague, buzzword-heavy language like 'comprehensive', 'AI-powered', and 'swarm coordination' without listing any concrete actions. It doesn't specify what the skill actually does (e.g., commenting on PRs, suggesting fixes, checking style). | 1 / 3 |
Completeness | The 'what' is vague (no specific actions listed beyond the abstract 'code review with swarm coordination'), and there is no 'when' clause or explicit trigger guidance at all. | 1 / 3 |
Trigger Term Quality | 'GitHub' and 'code review' are natural terms users might say, but the description lacks common variations like 'PR', 'pull request', 'review comments', 'diff', or 'merge request'. 'Swarm coordination' is technical jargon unlikely to be used by users. | 2 / 3 |
Distinctiveness Conflict Risk | 'GitHub code review' provides some domain specificity, but 'comprehensive' and 'AI-powered swarm coordination' are too vague to clearly distinguish this from other code review or GitHub-related skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
22%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill reads more like aspirational product documentation or a feature catalog for a hypothetical tool than a focused, actionable skill for Claude. It is extremely verbose with heavy redundancy, marketing-style formatting, and lacks the validation checkpoints critical for workflows involving PR modifications and auto-merging. The core value—teaching Claude how to perform code reviews—is buried under layers of configuration examples, analytics commands, and feature descriptions.
Suggestions
Reduce content by 70-80%: eliminate redundant command examples, marketing bullet points, emoji headers, and sections that describe features rather than instruct (e.g., monitoring/analytics, license, best practices checklists). Focus on one clear workflow.
Add explicit validation checkpoints to the core review workflow: verify gh auth status before starting, check command exit codes, validate review output before posting, and include error recovery steps (e.g., 'if review-init fails, check X').
Split detailed reference material (custom agent code, CI/CD YAML, configuration schemas, troubleshooting) into separate bundle files and reference them with one-line links from the main SKILL.md.
Remove or consolidate the 5+ near-identical 'review-init' command invocations into a single parameterized example with a brief table of available agent types and flags.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at 700+ lines with massive redundancy (the multi-agent review init command appears 3+ times), extensive marketing-style bullet points, emoji decoration, and sections explaining concepts Claude already knows. Much of the content is aspirational feature documentation rather than actionable instruction. The collapsible TOC, benefits lists, and best practices sections add significant bloat. | 1 / 3 |
Actionability | Commands are provided in bash syntax and appear concrete, but they reference a tool (ruv-swarm) with CLI flags that may or may not exist as documented. The webhook handler JavaScript is incomplete (missing body parsing), and many commands feel like hypothetical API designs rather than verified, executable instructions. The custom review agent code is more complete but still pseudocode-like in its check methods. | 2 / 3 |
Workflow Clarity | Despite being a multi-step process involving PR review, there are no explicit validation checkpoints or error recovery loops in the core workflows. The 'Complete Review Workflow' in Quick Start is just a sequence of commands with no verification that each step succeeded. The security agent section has a basic if/else but no feedback loop for fixing issues. For a skill involving destructive operations (posting reviews, requesting changes, auto-merging), the lack of validation caps this at 1. | 1 / 3 |
Progressive Disclosure | Uses HTML details/summary tags for collapsible sections which provides some structure, and references related skills at the bottom. However, with no bundle files provided, all content is monolithically packed into one enormous file. The TOC with internal anchors helps navigation but the sheer volume of inline content (config examples, workflow YAML, custom agent code, analytics commands) should be split into separate reference files. | 2 / 3 |
Total | 6 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (1141 lines); consider splitting into references/ and linking | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Repository
- ruvnet/ruvector
- Commit
c2089c4
- Reviewed
Table of Contents
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.