40+ integrations

Connect in 10 minutes. Stays out of your way.

Devloom reads from your existing observability pipeline — no agent rewrite, no vendor lock-in. If you already run Prometheus or OTEL, you're halfway there.

$ devloom connect --provider prometheus
[OK] Prometheus endpoint verified
$ devloom connect --provider loki
[OK] Loki log stream connected
$ devloom connect --provider jaeger
[OK] Trace ingestion active
[OK] Root-cause engine ready
Metrics
Prometheus

Pull or push metrics via remote_write. No scrape config changes needed.

Docs →
Grafana Mimir

Connect your Mimir cluster via remote_read. Historical and live metrics.

Docs →
Datadog (import)

Import Datadog metrics via API key. Read-only access — no agent change.

Docs →
CloudWatch

AWS CloudWatch metrics and alarms via IAM role cross-account access.

Docs →
Logs
Grafana Loki

Stream logs from Loki via LogQL. Label-based correlation with your metrics.

Docs →
Elasticsearch

Query log indices via ES API. Supports structured and unstructured logs.

Docs →
CloudWatch Logs

Subscribe to CloudWatch log groups. Lambda, ECS, EC2 — all supported.

Docs →
Traces
Jaeger

Connect to Jaeger collector API. Trace ingestion with full span metadata.

Docs →
Zipkin

Zipkin v2 JSON trace format supported. Drop-in for Zipkin-instrumented services.

Docs →
OpenTelemetry

OTLP native. gRPC port 4317 or HTTP port 4318. Zero config if you export OTLP.

Docs →
Alerting
PagerDuty

Receive PagerDuty alerts as Devloom incidents. Enrich on-call notifications with root cause.

Docs →
OpsGenie

Bidirectional OpsGenie integration. Acknowledge and resolve from Devloom UI.

Docs →
Slack

Post incident summaries with root cause to your on-call Slack channel. Configurable per severity.

Docs →
Deployment
GitHub Actions

Annotate deploys automatically via GitHub Actions workflow step. Links commits to incidents.

Docs →
GitLab CI

GitLab pipeline webhooks post deployment events to Devloom correlation model.

Docs →
ArgoCD

ArgoCD sync events annotate Devloom's causal model with exact deployment timestamps.

Docs →
First-class OpenTelemetry

OTEL-native from day one.

If your services export OTLP, Devloom ingests them without configuration changes. We follow the OpenTelemetry spec — OTLP/gRPC on port 4317, OTLP/HTTP on port 4318. No proprietary agent, no forked collector topology, no lock-in. If you later move to a different backend, your instrumentation stays unchanged. We're not trying to replace your existing observability stack — we sit alongside it, consuming the same telemetry signals.

$ OTELCOL_EXPORTER=otlp devloom listen --port 4317
[OK] OTLP gRPC listener started on :4317
[OK] Receiving spans from payments-service
[OK] Receiving spans from checkout-service
[OK] Causal model updated — 3 services correlated

Build custom context into root cause.

Standard OTEL traces capture span data. They don't capture why a deploy happened, which feature flag changed, or that a schema migration ran 3 minutes before the latency spike. Custom annotations let you inject that context directly into Devloom's causal model. The correlation engine uses annotations as high-weight signals: a deploy annotation that matches the incident timestamp window jumps immediately to the top of the root cause candidates list.

Python devloom-sdk
import devloom
 
# Annotate a deployment event with metadata
devloom.annotate("deploy", {
    "service": "payments-service",
    "version": "v2.4.1",
    "commit": "abc12f3",
    "config_changed": ["redis.pool_size"]
})
 
# Higher-confidence root cause when this correlates with an incident

SDK available for Python, Go, and Node.js. REST API for any other runtime.

Missing an integration? We add them fast.

We prioritize integration requests from users. Open a GitHub issue or email us directly — most requests ship within 2 weeks.