gemini
Use when the user asks to run Gemini CLI for code review, plan review, or big context (>200k) processing. Ideal for comprehensive analysis requiring large context windows. Uses Gemini 3 Pro by default for state-of-the-art reasoning and coding.
66
58%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./codex/gemini/SKILL.mdQuality
Discovery
67%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 has a clear 'Use when' clause and identifies the tool (Gemini CLI) and its primary use cases, which is good for completeness. However, the specific capabilities could be more concrete (what exactly does it do during code review?), and the trigger terms could better cover natural user language variations. The 'code review' and 'plan review' terms risk overlap with other review-focused skills.
Suggestions
Add more specific concrete actions beyond category names, e.g., 'Runs Gemini CLI to analyze codebases, generate review feedback, identify bugs, and process files exceeding 200k tokens'
Expand trigger terms with natural user variations like 'gemini', 'large file analysis', 'long document', 'big codebase review' to improve matching coverage
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (Gemini CLI) and some actions (code review, plan review, big context processing), but doesn't list concrete specific actions like 'analyzes diffs', 'generates review comments', etc. The actions are somewhat general categories rather than detailed capabilities. | 2 / 3 |
Completeness | Clearly answers both 'what' (run Gemini CLI for code review, plan review, large context processing using Gemini 3 Pro) and 'when' (explicit 'Use when' clause specifying user asks for code review, plan review, or big context processing). The trigger guidance is explicit. | 3 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'Gemini CLI', 'code review', 'plan review', and 'large context windows', but misses natural user variations like 'gemini', 'large file analysis', 'big file', '>200k tokens', or 'comprehensive review'. The term 'big context (>200k) processing' is somewhat technical. | 2 / 3 |
Distinctiveness Conflict Risk | 'Code review' and 'plan review' are fairly common tasks that could overlap with other code review or analysis skills. However, the specific mention of 'Gemini CLI' and '>200k' context processing helps narrow the niche somewhat. Could still conflict with general code review skills. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is highly actionable with excellent concrete examples and executable commands, but suffers severely from redundancy and verbosity. The same critical warnings and command patterns are repeated across multiple sections, inflating the token cost significantly. The content would benefit greatly from consolidation of repeated information and splitting reference material (model tables, troubleshooting) into separate files.
Suggestions
Consolidate the 'approval-mode yolo' warning to a single prominent section and reference it elsewhere instead of repeating it 6+ times throughout the document.
Remove marketing statistics (SWE-bench scores, Elo ratings, percentage improvements) from the model table — Claude doesn't need these to select the right model; just state the recommended use case.
Move the troubleshooting section and detailed model selection guide into separate referenced files (e.g., TROUBLESHOOTING.md, MODELS.md) to reduce the main skill's token footprint.
Eliminate the 'Common Use Cases' section entirely — it duplicates the examples already shown in 'Running a Task' with nearly identical commands.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose with massive redundancy. The hung process warning and fix appears 3+ times (in the critical warning section, troubleshooting section, and prevention section). The 'always use yolo for background tasks' instruction is repeated at least 6 times. The quick reference table, common use cases, and running a task section all repeat the same command patterns. Model details include marketing stats (SWE-bench scores, Elo ratings) that Claude doesn't need. The entire document could be cut by 60%+ without losing actionable information. | 1 / 3 |
Actionability | The skill provides fully executable, copy-paste ready bash commands for every use case. CLI flags are specific, examples include real command-line invocations with proper syntax, and the troubleshooting section has concrete diagnostic commands. | 3 / 3 |
Workflow Clarity | The 'Running a Task' section provides a clear numbered sequence, and there are validation steps for hung processes. However, there's no explicit validation checkpoint after running Gemini (e.g., checking exit code before proceeding), and the workflow for detecting/handling failures is scattered across multiple sections rather than integrated into the main workflow. | 2 / 3 |
Progressive Disclosure | The content is organized into sections with headers, but it's a monolithic document with significant content duplication that should be consolidated or split. The troubleshooting section, model selection guide, and common use cases could be separate reference files. No external file references are used despite the document being well over 100 lines. | 2 / 3 |
Total | 8 / 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
- jdrhyne/agent-skills
- Commit
6768672
- 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.