The nullplatform agent
The nullplatform agent is a lightweight service you can run anywhere, from your laptop to a Kubernetes cluster. It connects your environment securely to the nullplatform API, so you can automate tasks, monitor scopes and services, trigger actions, and handle event-based notifications—all without exposing your infrastructure.
This setup is privacy-first: your data stays in your cluster. Only minimal metadata and telemetry are sent to nullplatform.
How it works
The agent is a small binary that can run in many ways—on your local machine, in a container, or on a Kubernetes cluster (via Helm, Docker, or just the binary). Here's what it does:
- Authenticates with the nullplatform API using your API key
- Registers itself with optional tags (e.g.
environment:dev,team:secops) - Polls the API for matching tasks, scopes, or triggers
- Runs matching commands or workflows inside your environment
- Optionally sends logs, telemetry, and results back to nullplatform
🔒 The agent is outbound-only — you don’t need to open any firewall ports or expose your cluster.
Minimal flow
Key features at a glance
| Feature | Description |
|---|---|
| Tag-based routing | Agents only respond to tasks that match their tags |
| Secure API integration | Use API keys with fine-grained access |
| Notification channels | Run local scripts in response to platform events, using a git-audited flow |
| Flexible runtime options | Deploy the agent using Helm (K8s), Docker, binary, or serverless (e.g. Lambda) |
| Observability | Capture logs, exit codes, and outputs from workflows inside your infra. (Note: the agent itself doesn’t expose internal monitoring) |
| Multiple deployment models | Use in local dev (e.g. Minikube), cloud environments (e.g. EKS, GKE), or CI/CD pipelines |
What the agent unlocks
Once the agent is installed, it becomes the link between nullplatform and your infrastructure. Here’s what it lets you do:
-
Control your scopes directly in your environment
Launch tasks like security scans or provisioning workflows directly inside your infra. -
Connect platform events to your systems
Trigger commands in containers, virtual machines, or other systems when something happens on the platform. -
Use custom actions without extra glue code
Respond to events (like deployments or job completions) with local scripts—no brittle webhooks required. -
Manage internal services
Link the platform to services or APIs running inside your infrastructure. -
Send back telemetry and task results
Capture logs, outputs, and task status from your local runs and send them to the platform UI. -
Keep sensitive data local
Everything runs on-prem—no data leaves your infra unless you choose to share logs or outputs.
What’s next
Explore the following pages to learn more:
- Installation: Install the agent with Helm and verify it's working
- Deployment options: Run the agent via Helm, Docker, binary, or Lambda
- Notification channels: Let platform events trigger your local scripts
- Use cases: Examples of what the agent can do, including scopes, telemetry, automation, and compliance