https://github.com/podmortem/ai-interface

Backend service to interface with AI APis in the podmortem operator.
https://github.com/podmortem/ai-interface

ai devops failure-analysis incident-response kubernetes log-analysis mlops observability ollama openai operator pattern-matching pattern-recognition quarkus troubleshooting vllm

Last synced: 3 months ago
JSON representation

Backend service to interface with AI APis in the podmortem operator.

• Host: GitHub
• URL: https://github.com/podmortem/ai-interface
• Owner: podmortem
• License: apache-2.0
• Created: 2025-06-19T20:00:32.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-08-12T02:35:51.000Z (10 months ago)
• Last Synced: 2025-08-15T00:10:05.166Z (10 months ago)
• Topics: ai, devops, failure-analysis, incident-response, kubernetes, log-analysis, mlops, observability, ollama, openai, operator, pattern-matching, pattern-recognition, quarkus, troubleshooting, vllm
• Language: Java
• Homepage:
• Size: 95.7 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ai-interface

[![Security Rating](https://sonarcloud.io/api/project_badges/measure?project=podmortem_ai-interface&metric=security_rating)](https://sonarcloud.io/summary/new_code?id=podmortem_ai-interface)

> A Quarkus-based REST API service providing AI pod failure analysis for the Podmortem operator.

## Overview

This service acts as the primary interface for AI analysis operations, orchestrating interactions between the log parser results and various AI providers. It provides fault-tolerant endpoints for generating human-readable explanations of Kubernetes pod failures.

## REST Endpoints

- `POST /api/v1/analysis/analyze` - Analyze pod failures using AI providers

- `POST /api/v1/analysis/validate` - Validate AI provider configurations

- `GET /api/v1/analysis/providers` - List available AI providers

- `POST /api/v1/analysis/prompts/reload` - Reload prompt templates

- `GET /api/v1/analysis/prompts/status` - Get prompt template status

## Dependencies

- `common-lib` - Shared models and interfaces

- `ai-provider-lib` - AI provider implementations

## Building

```bash

./mvnw package

```

For native compilation:

```bash

./mvnw package -Dnative

```