CYBERSECURITY COPILOT

Meet Rant AI.

Use it in the web app, the Desktop App, or the rantai CLI. One agent that executes approved recon, audits, and IR tasks wherever you work.

Built by the CyberRant team. Trained on cybersecurity datasets, best practices, and open-source threat intelligence.

~ rantai

$ rantai chat

rantai > what changed in OpenSSH 9.6 from a security standpoint?

✓ summarizing CVE-2024-6387, advisories, exposure check…

$ rantai audit ./terraform

[agent] 2 critical, 5 high. open report? (y/n)

HOW YOU USE IT

Four surfaces. One agent.

In-platform chat

Ask Rant AI

Sits inside the CyberRant web app. Ask threat intel questions, decode logs, draft detection rules, and pull context from your own feed and DMs.

Native experience

Desktop App

The Rant AI Agent is built right into the CyberRant Desktop App. Sign in and start security operations with system tray access and auto updates.

Terminal power users

rantai CLI Agent

Install once, login once. The agent runs on your box, plans tasks, asks for approval, and executes only what you green-light.

On any post

@rantai mentions

Mention @rantai under a rant or comment. The bot replies inline with sourced, plain-text answers visible to the whole thread.

SETUP

Set up in under 60 seconds.

One install. One login. You're running.

01

Install the agent

Install from npm. The package installs locally, creates a Python venv, installs all requirements, and launches the runtime directly. Requires Node.js 18+. One-time setup.

npm install -g rantai
02

Run RantAI

Launches the interactive cybersecurity copilot shell.

rantai
03

If an old cached version exists

Delete the local cache, then run rantai again to reinitialise.

rm -rf ~/.rantai
04

If Python is missing — macOS

Install Python 3.12 via Homebrew, then re-run rantai.

brew install python@3.12
05

If Python is missing — Ubuntu / Debian

Install Python and required system dependencies, then re-run rantai.

sudo apt update && sudo apt install python3 python3-venv git -y
06

Verify installation

Confirm rantai is installed globally on your system.

npm list -g rantai
CAPABILITIES

What the agent can do.

Reconnaissance

  • Nmap and masscan scans with version detection
  • Subdomain, DNS, and ASN mapping
  • TLS and certificate inspection
  • Lateral-movement risk scoring

Threat Intelligence

  • CVE research with severity and exploitability
  • Vendor and CISA KEV lookup
  • IOC enrichment for IPs, domains, and hashes
  • Threat actor and campaign profiling

Log and Signal Analysis

  • Auth-log triage with geo enrichment
  • Suricata and Zeek pcap mapped to MITRE ATT&CK
  • Process-tree diff against a known-good baseline
  • Detection rule drafting

Code and Config Audits

  • Hardcoded secret scans
  • Weak-crypto, IDOR, and deserialization patterns
  • Dependency CVE mapping
  • Terraform, Kubernetes, and Dockerfile review

Incident Response

  • Timeline reconstruction
  • Containment playbook generation
  • Comms drafts for stakeholders
  • Post-mortem skeletons

Research and Drafting

  • Policy and runbook drafts
  • RFC summaries with risk callouts
  • Pentest report sections
  • Tabletop exercise scenarios
USAGE

Requests, resets, and top-ups.

Every plan ships with a monthly request budget shared across Ask Rant AI, @rantai mentions, and the CLI agent.

Free Forever

$0

20

requests / month

Rant AI Pro

$19/mo

500

requests / month

Rant AI Premium

$189/yr

2,000

requests / month

What counts as a request

Every Ask Rant AI message, every @rantai mention, and every CLI invocation that calls the agent. Cached follow-ups are free.

Monthly reset

Quotas refresh on the 1st of every calendar month. Your in-app dashboard always shows the next reset date.

Top up when you run out

Hit your limit before reset day? Buy a top-up pack from the Usage tab in settings and keep working without changing your plan.

CLI REFERENCE

Command cheat sheet.

rantai loginAuthenticate with CyberRant
rantai logoutClear local credentials
rantai chatOpen an interactive REPL
rantai scan <target>Recon sweep with approval gates
rantai triage <log>Analyze auth and system logs
rantai audit <path>Static review of code or IaC
rantai cve <id>Lookup with exploitability context
rantai ir <case>Spin up an IR playbook
rantai usageCheck remaining requests this month
SAFETY MODEL

Autonomy with brakes.

Designed so the agent never acts faster than your judgement.

Approval-gated execution

The agent never runs a command on your machine without an explicit y/n confirmation per action.

Local-first

The CLI executes on your box. Your files don't leave unless you ask the agent to fetch external context.

Scoped context

Retrieval covers your own feed and conversations only. The agent never trains on your data and never indexes other users'.

Run your first task.

Sign up, install the CLI, and let Rant AI take a real piece of work off your plate.