Logo of Akmatori
GitHub

Self-Hosted AI Incident Response for SRE, Platform, and DevOps Teams

Akmatori gives SRE and platform teams an AI incident responder that triages alerts, gathers evidence from your stack, and drafts safe next actions in minutes. Run it in your own infrastructure, connect the tools you already use, and keep humans in control with approval gates for production changes.

Start the 15-Minute POC
Open-source Apache 2.0Deploy with Docker Compose or HelmWorks with your existing alerts and runbooks

Start with a 15-minute proof of concept on your own alerts, then scale from a self-hosted pilot to managed infrastructure when your team is ready.

Get Started in 60 Seconds

Deploy Akmatori locally, install it on Kubernetes, or trigger your first AI investigation from the CLI. No vendor lock-in, full data control. 

# Clone and start Akmatori in under a minute
git clone https://github.com/akmatori/akmatori.git
cd akmatori
docker compose up -d

# Open http://localhost:8080 to access the UI
Quick start guide

Best for local evaluation and fast team demos

  • Boot the full stack on localhost with Docker Compose.
  • Open the UI immediately and connect your first alert source.
  • Validate incident workflows before touching production.
Prerequisites
Docker EngineDocker ComposeGit access
You are ready when
  • The web UI loads on localhost.
  • Core services show healthy in Docker Compose.
  • You can create a test incident from the dashboard.
Verify the stack is healthy
docker compose ps

You should see the Akmatori services in a running or healthy state before opening the UI.

Read Docker quick startView source on GitHub
Developer-First

Built for Engineers

Full REST API, webhooks, and native integrations. Connect Akmatori to your existing stack in minutes. 

# Install the Akmatori CLI
curl -fsSL https://get.akmatori.com | sh

# Authenticate with your instance
akmatori auth login --url https://akmatori.example.com

# List recent incidents
akmatori incidents list --status open --limit 5

# Trigger an investigation from the command line
akmatori incidents create \
  --title "High CPU on prod-web-01" \
  --severity critical \
  --source prometheus \
  --context '{"host":"prod-web-01","cpu":98.5}'

# Stream agent activity in real-time
akmatori agent logs --follow --incident inc_abc123

# Run ad-hoc diagnostics with the AI agent
akmatori agent run "Check why checkout-api has high latency"

# Export incident timeline to JSON for post-mortems
akmatori incidents export inc_abc123 --format json > postmortem.json
Powerful CLI
JWT & API Key Auth
OpenAPI 3.0 Spec
Python SDK
Webhooks
View Full API Reference
See It In Action

AI Agent Investigation Demo

Watch how Akmatori automatically triages an alert, runs diagnostics, and identifies root cause in seconds.

Click "Replay" to see the agent in action...
Skip to main content

What is Akmatori?

Akmatori is an advanced AI-powered agent designed to automate incident management, streamline troubleshooting, and optimize workflows for DevOps and SRE teams. By reducing downtime and boosting efficiency, it ensures seamless operations with intelligent, AI-driven solutions.

Key Features

Everything you need to automate incident management

Automated Incident Response

Resolve 80% of common alerts without human intervention. From detection to remediation in seconds, not hours.

Root Cause Analysis

Stop guessing. AI analyzes logs, metrics, and traces to pinpoint exactly why your service failed.

Alert Noise Reduction

Turn 200 alert storms into 1 actionable insight. Automatic deduplication and correlation across your stack.

Proactive Troubleshooting

Catch problems before your customers do. Pattern detection identifies anomalies early and suggests fixes.

Works With Your Stack

Prometheus, Kubernetes, Linux, PagerDuty, Slack, Datadog, and more. Plug into your existing toolchain.

Your Data, Your Servers

100% self-hosted. No telemetry, no external calls. Run with local LLMs for air-gapped environments.

Built for Engineers

No training required. Clean UI that shows what matters: incidents, runbooks, and agent activity.

15-Minute Setup

One docker compose command. Connect your alerts. Watch the AI learn your environment immediately.

Bring Any LLM

OpenAI GPT-5.4, Claude Opus 4.6, Gemini 2.5 Pro, OpenRouter, or your own endpoint. Swap providers anytime. No code changes, no lock-in.

Built for Your Reality

Real scenarios where Akmatori transforms how teams handle incidents

On-Call Engineer

Sleep Through the Night

Before

Woken at 3 AM for a disk space alert that just needed a log rotation

After

AI handles routine remediation. You only wake up for real outages.

SRE Team Lead

End Alert Fatigue

Before

Team burned out from 500+ weekly alerts, most are false positives

After

Intelligent correlation cuts noise by 90%. Engineers focus on real issues.

Platform Engineer

Kubernetes on Autopilot

Before

Hours spent debugging pod crashes and resource contention manually

After

AI diagnoses cluster issues, suggests fixes, and executes runbooks.

Startup Ops

Enterprise SRE Without the Headcount

Before

One person managing production with no budget for a full SRE team

After

AI multiplies your capacity. Ship features instead of firefighting.

Why Teams Replace Patches and Point Solutions

Akmatori is built for incident response, not just dashboards, docs, or chat answers.

Instead of Static Runbooks

Documentation goes stale fast, and engineers still need to translate alerts into the right recovery steps under pressure.

With Akmatori

Akmatori connects alerts to live context, picks the right runbook, and executes or suggests the next action automatically.

Instead of Generic AI Copilots

They can answer questions, but they are not wired into your incidents, permissions, tooling, or approval flow.

With Akmatori

Akmatori is built for operations work: alert ingestion, investigation, remediation, audit trails, and human handoff.

Instead of Legacy AIOps Suites

Expensive black boxes often force vendor lock-in and make self-hosting or local-model deployments painful.

With Akmatori

Akmatori stays open and flexible: self-host it, bring your own models, and integrate with the stack you already run.