gemini-review
Google Gemini CLI code review with Gemini 2.5 Pro, 1M token context, CI/CD integration
39
37%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/gemini-review/SKILL.mdQuality
Discovery
32%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 reads more like a product tagline or feature bullet point than a functional skill description. It lacks concrete actions (what specific tasks does it perform?) and entirely omits trigger guidance (when should Claude select this skill?). The mention of Gemini-specific details provides some distinctiveness but doesn't compensate for the missing structure.
Suggestions
Add explicit actions describing what the skill does, e.g., 'Reviews code changes using Google Gemini CLI, generates inline review comments, identifies bugs and security issues, and integrates findings into CI/CD pipelines.'
Add a 'Use when...' clause with natural trigger terms, e.g., 'Use when the user asks for code review via Gemini, wants to run Gemini CLI for PR feedback, or needs to set up automated code review in CI/CD.'
Replace the feature-spec style ('1M token context') with user-facing benefit language, e.g., 'supports reviewing large codebases up to 1M tokens' to clarify when this capability matters.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (code review) and mentions some specifics (Gemini 2.5 Pro, 1M token context, CI/CD integration), but these read more like feature specs than concrete actions. It doesn't list specific actions like 'reviews pull requests', 'generates review comments', or 'flags security issues'. | 2 / 3 |
Completeness | It partially addresses 'what' (code review with Gemini) but has no 'when' clause at all — no 'Use when...' or equivalent trigger guidance. Per the rubric, a missing 'Use when...' clause caps completeness at 2, and since the 'what' is also weak and reads like a feature list rather than a clear capability description, this scores a 1. | 1 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'code review', 'Gemini', 'CI/CD', but misses common user variations like 'review my code', 'pull request review', 'PR feedback', 'Gemini CLI'. The terms are somewhat technical and not fully aligned with natural user language. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'Google Gemini CLI' and 'Gemini 2.5 Pro' provides some distinctiveness from generic code review skills, but 'code review' and 'CI/CD integration' are broad enough to overlap with other code review or CI/CD-related skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable, executable content with complete CLI commands and CI/CD configurations, which is its primary strength. However, it is severely bloated with marketing-style content (benchmarks, competitive comparisons, feature/benefit tables) that wastes tokens without helping Claude perform code reviews. The lack of any bundle files means everything is crammed into one monolithic document with no progressive disclosure.
Suggestions
Remove the 'Why Gemini?' feature/benefit table, benchmark performance table, and 'Comparison: Claude vs Codex vs Gemini' section — these are marketing content that don't help Claude execute code reviews.
Split into multiple files: keep SKILL.md as a concise overview with quick-start, and move GitHub Action YAML, GitLab CI YAML, and authentication options into separate referenced files (e.g., GITHUB_ACTION.md, GITLAB_CI.md, AUTH.md).
Add validation steps to CI/CD workflows: verify CLI installation succeeded, check that review output is non-empty before posting, and handle error cases (e.g., rate limiting, auth failure).
Remove the installation prerequisites section explaining how to install Node.js — Claude knows how to install Node.js and this wastes tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~350+ lines. Includes unnecessary marketing content (benchmark tables, 'Why Gemini?' section), comparison tables with competitors, and explanatory text Claude already knows. The benchmark scores, feature/benefit tables, and competitive comparisons waste significant token budget without adding actionable value. | 1 / 3 |
Actionability | Provides fully executable code throughout: complete bash commands for installation, authentication setup, CI/CD YAML configurations for both GitHub Actions and GitLab, and concrete CLI usage examples. The GitHub Action workflow is copy-paste ready. | 3 / 3 |
Workflow Clarity | The CI/CD workflows show clear step sequences, but there are no validation checkpoints or error recovery steps. For example, the GitHub Action doesn't verify Gemini CLI installed correctly, doesn't handle review failures, and there's no feedback loop for when the review output is malformed or empty. | 2 / 3 |
Progressive Disclosure | Monolithic wall of text with everything inline — installation, authentication, interactive use, CI/CD for multiple platforms, configuration, comparison tables, and troubleshooting all in one file with no bundle files to offload content. The GitHub Action YAML alone takes ~40 lines that could be in a separate reference file. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- alinaqi/claude-bootstrap
- Commit
65efb33
- 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.