Which API keys are required for this skill?

You must set either a GEMINI_API_KEY or GOOGLE_API_KEY as an environment variable. These can be obtained from Google AI Studio.

Does this skill modify my source code?

No, the gemini-review skill is strictly advisory. It uses a mandatory --sandbox mode to ensure it only reads files and provides recommendations without making any changes to your project.

Can I use this for log files and documentation?

Yes, it is highly effective for analyzing large application logs, build outputs, technical specifications, and database schemas.

What happens if a review times out?

The skill includes built-in timeout recovery. If a review exceeds the limit, you can reduce the scope to specific directories or use the file-output pattern to capture results over a longer duration.

Why use Gemini for review when I'm already using Claude?

Gemini provides a massive 2-million-token context window, allowing it to see your entire codebase at once for architectural reviews, plus it offers an independent 'second opinion' to catch issues another model might miss.

Gemini Large Context Review

Name: Gemini Large Context Review
Author: nthplusio

bynthplusio

0•

安全与测试

Leverages Gemini's 2M token context window to perform deep, advisory-only reviews of large codebases, logs, and documentation within Claude Code.

The gemini-review skill integrates the Gemini CLI into your Claude Code environment to provide a high-capacity 'second opinion' for complex tasks. By utilizing Gemini's massive 2-million-token context window, it can ingest and analyze entire repositories, deep directory structures, or massive log files that exceed the practical limits of standard LLMs. Operating in a strictly read-only sandbox mode, it ensures that your source code remains untouched while providing structured, critical feedback on security, performance, and architectural integrity. It is the ideal tool for cross-module impact analysis, technical debt assessment, and identifying patterns across thousands of lines of code or logs.

主要功能

012-Million Token Context Window for analyzing massive files and directories.

02Advisory-Only Enforcement ensuring the AI never modifies source files directly.

03Multi-File Concatenation for cohesive, system-wide architectural reviews.

04Structured JSON Output support for machine-readable findings and automation.

05Configurable Timeouts to handle large-scale processing without hanging.

060 GitHub stars

使用场景

01Performing a full-repository security audit to find vulnerabilities across multiple interconnected modules.

02Analyzing thousands of lines of application logs to detect intermittent error patterns or performance regressions.

03Conducting migration impact analysis to see how a database schema change affects various parts of the codebase.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nthplusio/functional-claude gemini-review

For use in Claude.ai and ChatGPT

主要功能

012-Million Token Context Window for analyzing massive files and directories.

02Advisory-Only Enforcement ensuring the AI never modifies source files directly.

03Multi-File Concatenation for cohesive, system-wide architectural reviews.

04Structured JSON Output support for machine-readable findings and automation.

05Configurable Timeouts to handle large-scale processing without hanging.

060 GitHub stars

使用场景

01Performing a full-repository security audit to find vulnerabilities across multiple interconnected modules.

02Analyzing thousands of lines of application logs to detect intermittent error patterns or performance regressions.

03Conducting migration impact analysis to see how a database schema change affects various parts of the codebase.