Show HN：IncidentFox – 具備日誌採樣與 RAPTOR 檢索功能的開源 AI SRE

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

To see all available qualifiers, see our documentation.

AI-powered SRE platform for automated incident investigation

License

Uh oh!

There was an error while loading. Please reload this page.

incidentfox/incidentfox

Folders and files

Latest commit

History

Repository files navigation

IncidentFox

Our mission: Build the world's best AI SRE.

AI-powered incident investigation and infrastructure automation.

IncidentFox is an AI SRE / AI On-Call engineer that integrates with your observability stack, infrastructure, and collaboration tools to automatically investigate incidents, find root causes, and suggest fixes.

🌐 Try it live: ui.incidentfox.ai | 📧 Enterprise & On-Premise: [email protected]

Why IncidentFox?

✨ Key Features

Core Capabilities

Advanced AI Features

Enterprise Ready

Extensible & Customizable

🚀 Quick Start

Option 1: Local CLI (Fastest)

Try IncidentFox locally with an interactive terminal:

Full local setup guide: local/README.md

Option 2: Manual Setup

Environment Variables

🔌 Integrations

Slack Bot (Primary Interface)

Mention the bot in any channel to start an investigation:

Setup:

GitHub Bot

Comment on issues or PRs to trigger investigation:

Setup:

What happens:

PagerDuty (Auto-Investigation)

Automatically investigate when alerts fire:

Setup:

When an incident triggers, IncidentFox automatically:

A2A Protocol (Agent-to-Agent)

Allow other AI agents to call IncidentFox using Google's A2A protocol:

Supported methods: tasks/send, tasks/get, tasks/cancel, agent/authenticatedExtendedCard

REST API

Direct programmatic access:

🤖 Agent Architecture

Agent Capabilities

Tool Categories

📁 Repository Structure

🏗️ Deployment

IncidentFox can be deployed to any Kubernetes cluster. We provide Helm charts for Kubernetes deployment and Terraform modules for cloud infrastructure.

Prerequisites

Helm Chart Deployment

Helm Values Files:

See charts/incidentfox/README.md for full configuration options including OIDC, RBAC, and production hardening.

Terraform Infrastructure (Optional)

If you need to provision cloud infrastructure, Terraform modules are provided for each service:

Available Terraform modules:

Docker Compose (Development)

For local development without Kubernetes:

Architecture Overview

Service Endpoints (configure for your environment):

🔐 Security & Governance

📊 Web UI Features

Admin Console

Team Console

🧪 Testing

📈 Evaluation Framework

IncidentFox includes a comprehensive evaluation framework to measure agent performance on real incident scenarios. This enables continuous improvement and provides confidence in agent capabilities.

How It Works

Evaluation Scenarios

Scoring Dimensions

Total: 100 points per scenario

Latest Results

Example: Cart Crash Investigation

Scenario: Cart service pod is in CrashLoopBackOff

Agent Output:

Score: 90/100 ✅

Running Evals

Eval-Driven Development

We use evals to:

Target: ≥85 average score, <60s per scenario

🔒 Privacy & Telemetry

IncidentFox includes an optional telemetry system to help improve the product. Organizations have full control:

User Control

What We Collect (When Enabled)

What We DON'T Collect

Technical Implementation

See Telemetry System Documentation for complete details.

📚 Documentation

Architecture

Getting Started

Services

Advanced Features

🔗 Testing & Evaluation

For testing agent capabilities, we recommend:

OpenTelemetry Demo - Microservices demo app ideal for fault injection testing

Your own staging environment - Deploy IncidentFox alongside your existing staging Kubernetes cluster for realistic testing

🗺️ Roadmap

Completed

In Progress

💼 Commercial Options

IncidentFox is open source and free to use. For teams that need more, we offer:

Why On-Premise?

For organizations with strict security requirements:

Contact: [email protected]

Premium Services

These services are available separately for enhanced capabilities:

📄 License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

About

AI-powered SRE platform for automated incident investigation

Resources

License

Uh oh!

There was an error while loading. Please reload this page.

Stars

Watchers

Forks

Releases

Packages

  0

Contributors

  2

Uh oh!

There was an error while loading. Please reload this page.

Languages

Footer

Footer navigation