AI Build Troubleshooter

Relevant source files

The following files were used as context for generating this wiki page:

• src/ai/flows/ai-build-troubleshooting-assistant.ts
• src/components/dashboard/AITroubleshooter.tsx
• src/app/page.tsx
• src/ai/genkit.ts
• README.md
• src/ai/dev.ts

The AI Build Troubleshooter scans kernel compilation logs to identify GKI/KMI compatibility issues, failed patch hooks, and missing dependencies. It serves as an integrated diagnostic tool within the KernelForge dashboard to reduce debugging time for Android kernel developers.

Sources: README.md:12-14, src/components/dashboard/AITroubleshooter.tsx:47-48

System Architecture

The troubleshooter functions as a Genkit-powered AI flow integrated into a React frontend. While the production application exports as a static Android APK, the system includes a mock assistant for local execution without a backend server.

Component Logic

The UI component AITroubleshooter receives build logs as a string prop. It manages the analysis state and displays structured results containing a summary, identified errors, and suggested solutions.

Sources: src/components/dashboard/AITroubleshooter.tsx:10-28, src/ai/flows/ai-build-troubleshooting-assistant.ts:1-5

Data Flow

Sources: src/components/dashboard/AITroubleshooter.tsx:18-35, src/ai/flows/ai-build-troubleshooting-assistant.ts:25-30

Implementation Details

Data Structures

The system uses specific TypeScript types and schemas to ensure consistent communication between the frontend and the AI flow.

Data Type	Field	Type	Description
Input	buildLog	string	The raw text output from the kernel compilation process.
Input	deviceInfo	object	Optional metadata including model, codename, and GKI/KMI versions.
Output	summary	string	A high-level overview of the build status or identified issues.
Output	identifiedErrors	string[]	A list of specific error strings extracted from the logs.
Output	suggestedSolutions	string[]	Actionable steps to resolve the identified build failures.

Sources: src/ai/flows/ai-build-troubleshooting-assistant.ts:7-23

AI Configuration

The project utilizes Google GenAI via Genkit for its intelligence layer. In a non-mocked environment, it targets the gemini-2.5-flash model.

// Example Genkit Configuration
import {genkit} from 'genkit';
import {googleAI} from '@genkit-ai/google-genai';

export const ai = genkit({
  plugins: [googleAI()],
  model: 'googleai/gemini-2.5-flash',
});

Sources: src/ai/genkit.ts:1-7, src/ai/dev.ts:1-4

Execution Environment

The application behavior changes based on the deployment mode:

1. Static/Android APK Mode: The aiBuildTroubleshootingAssistant function returns a mock response. It informs the user that a remote AI endpoint is required for analysis within the APK.
2. Development Mode: The system can be started using Genkit CLI commands (npm run genkit:dev) to test flows against live AI models.

Sources: src/ai/flows/ai-build-troubleshooting-assistant.ts:32-47, package.json:7-8

Sequence Diagram

Conclusion

The AI Build Troubleshooter provides a structured interface for diagnosing kernel compilation failures. By translating raw terminal output into human-readable summaries and actionable solutions, it bridges the gap between complex log data and successful kernel deployment.