Skip to main content
The Observer Agent is a small process that runs in your network. It collects a value from a source, computes status against thresholds, and pushes the value and its verdict to Observer Cloud over an authenticated channel. It runs 15 source types:
  • Queries: Prometheus (PromQL), Loki (LogQL), Elasticsearch, AWS CloudWatch, and databases (PostgreSQL, MySQL, Redis, MongoDB).
  • Probes: HTTP (including mTLS), TCP, DNS, ping (ICMP), TLS certificate expiry, gRPC health, and WebSocket.
  • Push and local: OTLP (your services push to the agent), host vitals (CPU, memory, disk, network, load), and custom probes you register.
See Probe types for each type’s value and unit. Connection strings, certificates and tokens stay on the agent host: the cloud only stores the names of the environment variables that hold them. The agent is open source. Source at github.com/useobserver/agent, licensed Apache-2.0. The runtime is Bun on a distroless container image; the source is TypeScript.

​
Quickstart

Install from a release binary

Single-file executable from GitHub Releases. No runtime install on the host.

Install on Docker

Run the published image with three environment variables.

Install on Kubernetes

Deployment manifest with Secret-bound agent key.
Installing in an enterprise network with a proxy, TLS inspection, a private CA or restricted egress? See Run behind a proxy or in a restricted network.

​
Adjacent sections

  • Guides cover per-probe configuration, dashboard reading, key rotation, and diagnosis paths.
  • Reference lists every environment variable, every probe type, every dashboard panel, and the heartbeat payload shape.
  • Concepts covers the agent / cloud boundary, the local queue, and the design choices behind the runtime.
Where the cloud-side concerns liveDocumentation for status pages, SLOs, organisation setup, and the REST API is in the Documentation and API tabs. This tab covers the agent only.