skills.rest
EndUser123
EndUser123COMMUNITY
Explore all skills by this author →
GitHub

ai-pi-gg-gemma

Dispatch adversarial code review to Gemma 4-31B-IT via the pi agent and return JSON findings.

Updated Apr 26, 2026
Search Tags:#ai-assistant#code-review#multi-provider#adversarial-review#pi-agent#gemma
One-click install
npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-pi-gg-gemma
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-pi-gg-gemma
Source: https://github.com/EndUser123/cc-marketplace/tree/main/plugins/cc-skills-ai-cli/skills/ai-pi-gg-gemma
Command: npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-pi-gg-gemma

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables automated adversarial code review by dispatching to Gemma 4-31B-IT through the pi agent, producing structured, actionable findings to improve code safety and quality.

Core Features & Use Cases

  • Multi-provider dispatch: coordinates adversarial review with Gemma 4-31B-IT across providers.
  • Deterministic evaluation: returns a JSON object with score (0-1), summary (one sentence), and identified issues for the target.
  • Use Case: integrate into CI pipelines to systematically assess code quality and security during development and code review.

Quick Start

Run pi with the Gemma model to evaluate a target file and return a JSON object containing a score, a one-line summary, and any issues.

Frequently Asked Questions about ai-pi-gg-gemma

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate adversarial code review using Gemma?▼

Automated adversarial code review is handled by dispatching target files to the Gemma 4-31B-IT model through the pi agent, which returns structured findings including a score, summary, and identified issues.

What is adversarial code review and how does it evaluate code quality?▼

Adversarial code review evaluates software across files and repositories to produce deterministic scoring and issue reporting. It systematically assesses code safety and quality by returning a 0-1 score, a one-sentence summary, and a list of identified issues.

Do I need the pi agent and Gemma model access to run automated code reviews?▼

Yes, running automated code reviews requires the pi agent setup and Gemma model access. The Skill coordinates multi-provider dispatch to Gemma 4-31B-IT to generate the structured JSON output.

What's the best way to integrate adversarial code review into a CI pipeline?▼

Integrating adversarial code review into CI pipelines is achieved by running the pi agent with the Gemma model to evaluate target files during development. It returns a deterministic JSON object containing a score, summary, and issues for downstream automation.

What format does the Gemma code review output use for downstream automation?▼

The code review output uses a JSON format containing a score from 0 to 1, a one-sentence summary, and an array of identified issues. This structured format enables direct consumption by downstream automation systems.

Related Skills

Explore More Software Engineering Skills→
EndUser123EndUser123

ai-pi-zai-glm51

Automate adversarial code review with GLM-5.1 via pi and report score, issues, and synthesis.

Community
Advanced
OmniNode-aiOmniNode-ai
3

pr-review-dev

Consolidate PR comments and CI failures, then dispatch agents to fix issues.

Official
Advanced
EndUser123EndUser123

ai-pi-nv-qwen-coder

Dispatch adversarial code review to Qwen3 Coder 480B via pi agent.

Community
Advanced
openeuler-mirroropeneuler-mirror
3

seagate-farm-disktool-health-analysis

Analyze Seagate FARM telemetry logs for head-level disk health diagnostics.

Official
Advanced