The governed platform you don’t have to build.
Run any agent, govern every action, watch the meter, pause anytime. The security engineering is already done and kept on.
Reply within one business day · priced per org · bring your own keys, never marked up
What does the agen-x platform include?
agen-x is a managed, governed platform for AI agents. It includes 15 runtimes (OpenClaw, Claude Code, Codex, Hermes supported; 11 available), verifiable workspace isolation, strict network boundaries with allowlists, scoped credentials rotatable live, ask-a-human interrupts for any runtime, real-time metering with hard caps and pause, native SSO and RBAC, full audit trails with export, each runtime’s native UI behind your SSO, and templates, channels, and orchestration for the agents you add next.
- OpenClaw
- Claude Code
- Codex
- Hermes
- GitHub Copilot
- Goose
- OpenCode
- Gemini CLI
- Cursor
- Qwen
- Kimi
- DeepSeek
- ACP
15 runtimes. One policy for every agent your team runs.
Run the agent your team already uses
15 runtimes. Supported: OpenClaw, Claude Code, Codex, Hermes. Available: GitHub Copilot, Goose, OpenCode, Gemini CLI, Cursor, Qwen, Kimi, DeepSeek, and ACP agents. Paying teams choose what moves up.
- One policy for every agent your team runs
Isolation your security engineer can verify
Workspace isolation your security engineer can verify, enforced on every agent automatically.
- True tenant and agent isolation by default
Strict network boundaries
Agents reach only the domains you allow. Everything else is denied and logged.
- No exfiltration paths. No stealer payloads.
Ask a human first
Any agent can ask a person before a risky step, not only the runtimes that ship their own approvals.
- Irreversible actions wait for explicit approval
Your subscription or your keys
Claude Code, Codex, and Copilot on your own plan. Everything else on your own keys. Credentials scoped per agent, rotatable live.
- Keep your negotiated rates. Zero orchestration fees. No plaintext leaks.
A meter you can see — and stop
Real-time billing dashboards with hard budget caps.
- No bill shock. Paused agents cost disk only.
One security review
Native SSO, robust RBAC, and full audit trails with export.
- Security approves the platform once, not each agent
Native UI, behind your SSO
Runtime control UIs served through an authenticated proxy.
- No exposed listeners or separate logins
A workforce on the same platform
Templates, channels, and orchestration for the agents you add next. Same policy, same meter, same bill.
- No new vendors. No new security reviews.
How it works
Pick a runtime.
Bring your own keys.
Set the fence.
Name the allowed network domains.
Let it run.
Metered by the hour. Pause anytime.
Watch the fence hold.
Tell us what you want to run. We reply within one business day.
Reply within one business day · priced per org · bring your own keys, never marked up
