/observe
GitHubGet started

Observability workspacefor humans and agents.

Ask for a dashboard in a sentence and get one that works, on the Prometheus, ClickHouse, Loki and Tempo you already run. Every query runs before it is saved. The dashboards you built in Grafana come along.

$curl -fsSL https://slashobserve.com/install.sh | sh

Installs the observe CLI on macOS and Linux. The server is one binary beside Postgres; it runs in your network and sends only the agent's tool results to the model you configure.

observe.example.com/b/7c2e91a4-5b0d-4e8f-9a1c-3d6f2b8e4a17
/Dashboards/Checkout health
Overview
Requests/s
1.9kreq/s
Error ratio
0.4% OK
p95 latency
212ms
Pods ready
3
Traffic
Requests by pod
checkout-0checkout-1checkout-2
Latency by route
/checkout/cart
Errors by route
routeerrorsratio/checkout410.9%/cart30.1%/product/{id}00.0%/search10.0%
Internals
DB pool in use
in use
Queue depth
18
Payment success
99.1% OK
Reservations/s
64req/s
Cache hit ratio
94%
checkout health by pod, last 6h
Ran 15 steps in 38s
Checkout health12 panels from promOpen dashboard
Saved as version 1, just now
Request 7c2e91a4
Ask for another change to Checkout health
Each change is saved as a new version of the dashboard.

A build, played back.

Your Grafana dashboards come along.

Their folders, variables and links come too. Panels the importer cannot match go to the repair queue, a failing query can get a fix from the agent, and every old link keeps resolving.

Coming from Grafana →
observe.example.com/import/repair
Repair queueAll foldersRepair all
DashboardFolderUnverifiedPlaceholdersImportedRepairActions
Kubernetes / PodsPlatform / Kubernetes3 panels02 h agoRepairingRepair with the agent
Node exporter fullInfra5 panels, 1 variable02 h agoRepair queuedRepair with the agent
Payments latencyPlatform / Shop1 panel02 h agoRepaired, still failing12 min agoRepair with the agent
Ingest pipelineObservability2 panels32 h agoRepair failed1 h agoRepair with the agent
Checkout servicePlatform / Shop1 panel02 h agoNot requestedRepair with the agent
Graphite legacyInfraNone42 h agoNot requestedRepair with the agent
Old links
grafana.internal/d/rYdNrFH/node-exporter-fullobserve.internal/d/rYdNrFH/node-exporter-full302 to /d/9c41e7b2-0d5a-4f3e-b8a1-6e2c7d4f1a90/node-exporter-full same dashboard

Metrics, logs and traces, on the stores you already run.

Prometheus, ClickHouse, Loki and Tempo. A log line opens its trace.

/d/4f1c8a2e-93b7-4d1a-b6e5-0c2f7d9a1e34/checkout-health
Dashboards/Shop/Checkout healthLive · v7 ▾Controls
p95 latency by routeprom
Payment errorsloki
/logs?ds=loki&q={namespace="shop"}
LogsStop live tail
2026-11-06 14:02:11.204Errorpayment declined order=8812
2026-11-06 14:02:11.198InfoGET /cart 200 12ms
2026-11-06 14:02:10.911Warningdb pool near max 19/20
2026-11-06 14:02:09.540Errorupstream timeout after=2.0s
2026-11-06 14:02:09.102InfoGET /product/41 200 8ms
2026-11-06 14:02:08.990Inforeserve ok sku=41 qty=1
2026-11-06 14:02:08.421Errorpayment declined order=8809
2026-11-06 14:02:08.106Infoorder processed order=8807
2026-11-06 14:02:07.873InfoGET /cart 200 9ms
/t/7f3a9c1e5b2d4f6a8c0e1d3b5a7f9c2e?ds=tempo
frontend POST /checkout2026-11-06 14:02:09.1122.31 s
frontend POST /checkout
checkout place_order
inventory reserve
checkout db.query
payments POST /charge
http.request.methodPOSThttp.response.status_code504server.addressgateway.psp.example
payments authorize_card
frontend render

Signals →

In the browser, the terminal, or your own agent.

The web app, the observe CLI, and MCP for Claude Code and other MCP clients. Same save checks everywhere.

observe.example.com/c/3b9d01c2-6a4e-4f0b-8d21-9e7c5a1f4b60
Add errors by route and pin the p95 panel on top.
Ran 6 steps in 14s
  1. Discoveringlist_metricsListed 3 metrics on prom
  2. Checking querieslint_queryChecked a query on prom: 0 errors
  3. Running queriesquery_rangeRan a query on prom: 4 series
  4. Savingupdate_panels 3Saved panel 13: Errors by route (row Traffic)Moved 1 panel in the build (13 in all)Saved version 7 (3 rows, 13 panels)
p95 latency by route
Errors by route
Checkout health13 panels from promOpen dashboard
Saved as version 7, just now
Request 3b9d01c2
Checkout healthAsk for another change to Checkout healthchlokipromtempo
Each change is saved as a new version of the dashboard.
zsh
$ observe "queue health, last 6h"Panel 1 of 7: Queue depth (timeseries)Panel 2 of 7: Oldest order age (stat)Panel 3 of 7: Workers busy (timeseries)…Saved version 1 https://observe.example.com/d/9b2e/queue-health $ observe logs '{app="orders"} |= "retry"' --datasource loki --tail
Claude Code · mcp: slashobserve
› build a dashboard for the payments service ⚙︎ list_datasources⚙︎ list_metrics scope={job="payments"} 9 families · 2 runtime⚙︎ lint_query · query_range ×9⚙︎ create_dashboard saved /d/c41a/payments · uncovered: []
Apache-2.0Open source, no open core tricks
One binaryPlus Postgres. Helm chart and compose included
Your networkRuns in your network. Only the agent's tool results reach the model you configure.
Any modelAnthropic, OpenAI-compatible, or local

Runs on AWS
with Docker and Postgres.

git clone https://github.com/slashobserve/slashobserve && cd slashobserve && make demo

Try it on a laptop first: make demo brings up a shop, Prometheus, ClickHouse, Loki, Tempo and three reference dashboards in about two minutes. Add a model key when you want the agent.

Read the quick start