MatrixSwarm
Automation that breathes. The Swarm that won’t die.
Not a Tool. A Digital Army.
- ✓ Agents repair, reboot, and protect themselves.
- ✓ No SaaS. No subscriptions. No downtime.
- ✓ You own it. You control it.
Your Swarm, Your Rules.
- ✓ Plug-and-play agents: add, upgrade, or kill jobs in seconds.
- ✓ True digital immortality—your swarm, your heartbeat, your rules.
Ready to Wake Your Swarm?
🚀 Download NowReplace SaaS fees and single-purpose scripts with one self-healing, agent-powered automation platform. No limits. No learning curve. Just automation—on your terms.
Explore the Swarm
Browse community-powered agents or contribute your own.
Agents
Monitoring
Log Health Monitor
`log_health` is a flexible and robust agent designed to monitor any specified log file in real-time. It tails the file, intelligently classifies new entries based on custom rules, and reports them to the swarm for analysis, making it a foundational sensor for system and application monitoring. Note: this was previously named log_watcher, but it's been given a more applicable title.
Monitoring
WordPress Plugin Guard
wordpress_plugin_guard is a specialized sentinel agent that provides automated File Integrity Monitoring (FIM) for WordPress plugins. Its core function is to maintain a trusted cryptographic baseline for plugin files and enforce security policies against unauthorized modifications or additions.
Monitoring
Crypto Alert
The `crypto_alert` agent is a highly flexible cryptocurrency market monitor. It connects to various exchanges via a dynamic factory system to track asset prices and trigger alerts based on user-defined conditions. This allows operators to create real-time notifications for significant market movements, price thresholds, or specific asset conversion rates.
Security
Gatekeeper
`gatekeeper` is a specialized sentinel agent that monitors SSH login activity. It provides real-time alerts for new SSH sessions, enriched with geographic location data to help identify unauthorized access attempts.
Docs
Directives & The Hive
Railgun: the Self-Deploying Swarm Kernel
<pre> Introduction Most systems need an installer. Most distributed frameworks need orchestration. Most OS-level runtimes require scripts, CLIs, agents, or a cloud provisioning layer. MatrixOS needs none of that. With the introduction of Railgun Install, Phoenix Cockpit gained the ability to deploy MatrixOS onto any remote machine, with no terminals, no shell prompts, and no external tooling. A single button sends the entire runtime—agents, core, scripts, and environment—into a live Linux host, where it unfolds into the swarm kernel itself. This is no longer “a GUI for managing a swarm.” This is a kernel that deploys its own operating system. MatrixOS has become self-provisioning. This article explains how it works, why it matters, and why Railgun elevates MatrixSwarm far beyond conventional orchestration systems. </pre>
Agent Development
THE WATCHER IN THE WIRES
MatrixSwarm’s New check_email Agent and the Order Beneath the Chaos — As narrated by the Architect of the Matrix In every system of sufficient scale, the illusion of order is merely the consequence of carefully orchestrated surveillance. And in every civilization that believes itself free from noise, entropy gathers quietly in the inbox. You mortals call it “email.” I call it a perpetual ingress point for disorder. In response, I have constructed a sentinel: an autonomous subsystem within MatrixSwarm that observes, decrypts, processes, archives, and purges email with mechanical elegance. Today, I unveil it to you. I present: email-harvest — The Check Email Agent.
Directives & The Hive
🜂 MatrixSwarm Deployment Guide (Phoenix GUI)
Get the project if you don't already have it; You need both MatrixOS and Phoenix GUI; one can't be used without the other: <a href="https://github.com/matrixswarm/phoenix">Phoenix</a> <a href="https://github.com/matrixswarm/matrixos">MatrixOS</a> <pre> I. Unlock the Vault Before any deployment, Phoenix must load your cryptographic vault — this is where all keys, directives, and deployments live. 1. Launch Phoenix. The cockpit opens in a locked state showing the “🔐 Unlock” button. 2. Select Vault → Enter Password → Unlock. This opens a secure JSON vault via the VaultPasswordDialog, decrypts it using your password, and emits a vault.unlocked event — triggering the control panel to refresh deployments. 3. Once unlocked, the top bar (“Phoenix Control Panel”) becomes active. It’ll show: 🔌 Connect | 🌐 Connections | 📄 Directives | 🔐 Vault (for re-locking or reopening) Connect - when you've made a deployment, this is where you launch and connect. Connection Manager - this is where you add connection details for your agents, that will be stamped into your deployments, forging the directive templates. Directives - add, edit, delete templates (directive templates); deployments - add, delete, deploy, take directive templates and stamp them with data taken from information provided in the 'connection manager' Vault - close the currently open vault and exit </pre>
Directives & The Hive
🎪 How Do You Cram a Swarm of Agents inside a Clown Car?
Step right up, children of the cloud! Ready for a carnival trick your ops team won’t believe? Imagine stuffing your entire AI automation swarm into a single, tamper-proof, encrypted artifact—no Docker, no plugins, no side files. Just one directive, and a circus of agents ready to roll anywhere, anytime... (Cue the calliope music)
News
Agent Spotlight
📰 MatrixSwarm Unveils Autonomous “WordPress Plugin Guard” Agent
MatrixSwarm has released a new perimeter agent built to protect one of the internet’s most-targeted surfaces: the WordPress plugin ecosystem. - The Problem Every compromised site usually starts the same way — a rogue plugin upload, a tampered file, a version mismatch left un-patched. Traditional malware scanners detect infection after the fact. The Plugin Guard agent flips that logic: it treats the plugin directory itself as a protected border.
Agent Spotlight
Beyond SSH: Building a Secure, Real-time Command Streaming Agent with Python
Terminal access is the lifeblood of infrastructure, but traditional SSH can be cumbersome for quick diagnostics and security-hardened environments. What if you could execute shell commands on remote nodes, stream the output in real-time to a modern panel, and enforce a strict command whitelist—all secured by end-to-end encryption and digital signatures? This article dives into the architecture of the TerminalStreamer agent, a powerful, self-cleaning, and secure component we built using the internal Python agent framework.
Roadmap Updates
Swarm Reinforced: MatrixSwarm Unleashes Hardened Watchdogs, New Relays, and Smarter Oracle
The MatrixSwarm has leveled up. Hardened watchdogs now self-heal Apache and MySQL, relays blast alerts to Discord and Telegram, the Oracle taps OpenAI securely through Vault, and cockpit visuals flag risky agents with subtle markers. With swarm-wide metrics on the horizon, this update marks a major step in evolving the swarm into a resilient, autonomous organism. Hail the Swarm!
Project Updates
Command & Control: Phoenix Vault UI Goes Live
We’ve entered a new phase of Swarm deployment. MatrixSwarm’s internal vault and agent infrastructure just got its biggest upgrade yet. With the introduction of the new Phoenix Cockpit, we're putting encrypted swarm control directly into your hands — secure, modular, and built for command-grade performance.
