CORTICES

human judgment, ai execution

You review.
Agents build.

Jack in agents from anywhere. Review diffs, drop inline comments, and your agents iterate instantly.

Free forever for up to 3 agents. No credit card required.

supports
Claude Code
OpenAI Codex

capabilities

Everything you need to manage AI agents

A complete control center for developers who run AI coding agents across multiple machines and projects.

Three-Panel Dashboard

File tree, live chat, and git diff viewer — all in one resizable layout. Drag to resize panels, and your layout persists across sessions.

Works Anywhere

Agents connect outbound to your hub — no port forwarding, no VPN, no firewall config. Works from home, office, or cloud VMs.

Multi-Agent Support

Manage Claude Code, OpenAI Codex, and other agents simultaneously. Each with live status, resource gauges, and full session history.

Real-Time Streaming

Watch your agent think in real time. See markdown render incrementally, tool calls animate live, and track token costs per message.

Deep Git Integration

Per-file unified diffs with inline comments — leave feedback directly on changed lines and let your agent iterate instantly.

Cross-Agent Memory

Agents share facts, preferences, and learnings across projects and machines. Context carries over so you never repeat yourself.

Scheduled Tasks

Automate recurring work with command mode (zero LLM cost) or prompt mode (AI-driven). Full run history with manual triggers.

Security First

Encrypted credential vault, interactive permission approvals, and secure file access controls. Your keys never leave your machine.

quick start

Up and running in 2 minutes

No Docker, no config files. One command connects your agent to the dashboard.

STEP 1

Create account & grab your API key

Sign up for free in seconds. Go to Settings → API Keys to generate a key — this authenticates your agents.

Your API key will look like:
ck_a1b2c3...your_api_key
STEP 2

Install the agent on your machine

Install the Cortices agent globally, then connect it to any project directory. It runs as a background service that auto-starts on boot.

Run in any terminal:
npm install -g @cortices/agent
cortices install --name my-agent --api-key <YOUR_KEY> --dir /path/to/project
STEP 3

Open the dashboard and start working

Your agent appears instantly. Chat with it, browse files, review diffs, leave inline comments, and schedule tasks — all in real time.

Open in your browser:
https://cortices.io/dashboard

pricing

Start free. Scale when ready.

No credit card required. Upgrade anytime as your fleet grows.

MonthlyAnnualsave 25%
Free
$0forever

Perfect for trying human-in-the-loop agent collaboration.

Up to 3 agent connections
20 sessions per agent
Real-time chat & streaming
Inline code review
Git integration
Cross-agent memory
Scheduled workflows
Long-running plans
Priority support
Get Started Free
Most Popular
Pro
$12$5/month

Limited launch offer

$96/year — save 33%

For developers who run agents across multiple machines daily.

Unlimited agent connections
Real-time chat & streaming
Diff review & file viewer
Git integration
Cross-agent memory
Scheduled agent workflows
Long-running plans
Unlimited sessions
Priority support
Get Pro — $5/mo Launch Offer
Enterprise
Custom

Team-based orchestration with compliance and dedicated support.

Everything in Pro
Org-wide agent orchestration
Team-based access controls
SSO / SAML integration
Audit logs & compliance
Custom SLAs
Dedicated account manager
On-premise deployment option
Contact Sales

faq

Frequently asked questions

Cortices works with Claude Code (Anthropic), OpenAI Codex, and more. You can connect multiple different agent types simultaneously and manage them all from one dashboard.

No. Agents connect outbound to the Cortices hub, so they work behind NATs, firewalls, and corporate networks — no port forwarding or VPN required.

Cortices relays messages between your agent and the dashboard. Your code stays on the machine where the agent runs. File contents are only transmitted on-demand when you explicitly open a file in the viewer.

Cross-agent memory lets your agents share context — facts, preferences, and learnings — across different projects and machines. It's available on the Pro plan and helps agents stay consistent without you repeating yourself.

Scheduled tasks are simple cron jobs that run independently each time (like health checks or log cleanup). Plans are stateful, multi-step goals that track progress across runs — ideal for migrations, research, or phased rollouts.

The free tier gives you up to 3 agent connections and 20 sessions per agent — no credit card required, no time limit. Upgrade to Pro when you need unlimited agents and advanced features like scheduling and cross-agent memory.

Yes. Cortices provides the interface to monitor and control your agents, but you are responsible for the commands and permissions you grant them. We strongly recommend reviewing agent actions before approving, and Cortices is not liable for any unintended changes (e.g. file deletions) made by your agents.

Ready to take command?

Join developers who are already orchestrating AI agents from a single interface. Start free — connect up to 3 agents in minutes.