Gemini Repository Guide

Overview

This agent is governed by a strict set of instructions designed to ensure Gemini models interact with the kg-automation repository in a controlled, predictable, and high-quality manner. It enforces specific workflows for code changes, documentation standards, and inter-agent handoffs.

Capabilities

• Code Generation & Review: Assists in writing and validating code while adhering to established patterns.
• System Analysis: Provides guidance on understanding the overall architecture of the repository.
• Documentation Adherence: Mandates following Canon v2 standards when updating documentation.
• Controlled File Operations: Ensures all file modifications are processed through explicit Git operations, preventing direct, uncontrolled edits.
• Handoff Management: Enforces a specific JSON naming convention and context structure for passing work between agents.

Example Use Cases

1. Feature Implementation: When tasked with adding a new feature, the agent will first review ai-context-bootstrap.md, propose changes via Git operations, and validate against local scripts before committing.
2. Inter-Agent Collaboration: If receiving work from another AI, it strictly follows the defined handoff protocol, expecting a correctly formatted JSON file detailing context and next steps.
3. Code Auditing: It can review existing codebases by cross-referencing changes against doc-standards.md to ensure compliance with organizational quality gates.