AutoMend
AI-Powered Infrastructure

Your systems run
themselves

Autonomous monitoring, diagnostics, and remediation — your infrastructure stays healthy without constant human attention.

-- Health Score
-- Tasks Today
-- Uptime

Everything you need to
forget about maintenance

Proactive system care that catches issues before they become problems.

Autonomous Monitoring

CPU, memory, disk, and network tracked 24/7. Anomaly detection catches problems before they escalate.

Intelligent Diagnostics

Deep system analysis with actionable recommendations. Know exactly what needs attention and why.

Auto-Remediation

Configurable automatic fixes. Set it and forget it — or review every action before it runs.

Smart Alerting

Severity-based alerts with multi-channel delivery. Get paged for critical issues, emailed for warnings.

Real-Time Dashboard

Beautiful web dashboard with live metrics, charts, and system health at a glance. Dark mode included.

Scheduled Maintenance

Automated log rotation, disk cleanup, security scans, and backups. CI/CD via GitHub Actions.

Crypto Treasury

Multi-chain portfolio tracking, PnL monitoring, and token search powered by Bankr. All in the dashboard.

From zero to
self-healing infrastructure

Three steps to never worrying about routine maintenance again.

01

Deploy the Agent

Install via npm, configure your environment, and start the agent. Works on any Node.js server.

# One command to start npm start
02

Agent Monitors & Learns

System metrics are collected every 30s. The diagnostic engine builds a baseline of normal behavior.

# Health check API curl https://automend.vercel.app/api/overview
03

Auto-Heals & Reports

Issues are detected, diagnosed, and remediated automatically. You get a daily report via dashboard.

# Dashboard https://automend.vercel.app

See AutoMend
in action

Watch the dashboard live — monitoring, diagnostics, and AI-powered chat in under 2 minutes.

automend-demo.mov

How AutoMend works
under the hood

A modular agent architecture that monitors, diagnoses, and remediates — all with configurable safety boundaries.

System Monitor

Collects CPU, memory, disk, and network metrics every 30s. Detects anomalies against configurable thresholds.

CPU Memory Disk Network

Diagnostic Engine

Analyzes metrics against baselines, identifies root causes, and recommends remediation actions.

Health score · Alert correlation · Root cause analysis

Remediation Engine

Executes fixes with configurable safety — auto-approve, require approval, or dry-run only for destructive actions.

Auto Approval Dry-run

Built with modern,
battle-tested tools

From Node.js orchestration to AI-powered multi-agent systems.

Node.js

Core agent orchestration, monitoring loops, and event-driven architecture.

Python / Flask

Dashboard server, REST API endpoints, and WebSocket streaming.

CrewAI

Multi-agent AI orchestration — Senior Researcher, Coder, and Reviewer agents working together.

Chart.js

Real-time interactive charts for system metrics, health scores, and performance analytics.

GitHub Actions

CI/CD pipeline runs daily maintenance at 3 AM UTC. Automated tests and deployment.

Vercel

Serverless Flask deployment with Python runtime. Global edge network for low-latency API access.

Get AutoMend running
in under a minute

One command to start monitoring. A few more to unlock the full stack.

1

Install & Start

git clone https://github.com/kenGucci/automatic-maintenance.git cd automatic-maintenance npm install npm start
2

Launch Dashboard

pip install flask gunicorn python3 dashboard/app.py

Dashboard at http://localhost:8080

3

Configure

cp .env.example .env # Edit thresholds, notifications # and safety boundaries

Set AUTO_REMEDIATION=false for read-only mode

Built in public,
for the community

AutoMend is open source and free to use. Contributions, issues, and stars are welcome.

Powered by CrewAI
orchestration

Three specialized AI agents working together to keep your systems healthy.

Senior Researcher

Researches deployment processes, best practices, and emerging patterns. Keeps your infra current.

47 Tasks Done
94% Success Rate

Senior Coder

Writes and fixes maintenance code on the fly. Patches vulnerabilities and optimizes performance.

63 Tasks Done
91% Success Rate

Code Reviewer

Reviews all code changes for bugs, security issues, and style violations before deployment.

55 Tasks Done
97% Success Rate

Questions?
We have answers.

Everything you need to know about AutoMend.

What does AutoMend actually do?
AutoMend monitors your system metrics (CPU, memory, disk, network), runs diagnostics when thresholds are crossed, and automatically remediates common issues like log rotation, disk cleanup, and cache flushing. It sends you a report of everything it did.
Is it safe? Can it break my system?
AutoMend is built with safety first. Destructive actions require approval by default. There's a read-only mode (set AUTO_REMEDIATION=false), dry-run sandboxing, an audit trail, and rollback capability. You control the safety boundaries.
Do I need to replace my existing monitoring stack?
No. AutoMend sits on top of PagerDuty, Grafana, Datadog, and your existing runbooks. It reads what your monitoring already sees and acts on it — closing the gap between alert and resolution.
What tech stack does it require?
Node.js for the agent, Python/Flask for the dashboard. Optional: CrewAI for multi-agent orchestration. The dashboard deploys to Vercel. Works on any Linux/macOS server that supports Node.js 18+.
Is AutoMend free to use?
Yes. AutoMend is open source under the ISC license and completely free to use. Self-host it on your own infrastructure with no restrictions.
How do I contribute or report issues?
Star the repo on GitHub, open issues for bugs or feature requests, and submit PRs. Follow @Auto_Mend on X for updates. All contributions are welcome.

Ready to let your infrastructure take care of itself?

Deploy the agent in one command, then watch the dashboard light up. No credit card required.

A
AutoMend Assistant
Online