接入你的智能体
從發布者安裝。不是 PlugMCP 端點。
遠端 URL 還是本機指令? 兩種安裝方式的差異
obs-mcp is an MCP server that lets LLMs query Prometheus or Thanos Querier and Alertmanager in Kubernetes. It optionally supports Loki for logs, Grafana Tempo for traces, and OpenTelemetry Collector configuration assistance. Enable additional toolsets with --toolsets (e.g., --toolsets observability/metrics,observability/logs,observability/traces,observability/otelcol).
[!NOTE] This project is moved from jhadvig/genie-plugin preserving the history of commits.
Run make help to see all available commands.
The easiest way to get the obs-mcp connected to the cluster is via a kubeconfig:
make run
Or directly:
go run ./cmd/obs-mcp/ --listen 127.0.0.1:9100 --auth-mode kubeconfig --insecure
This will auto-discover the metrics backend in OpenShift. By default, it tries thanos-querier route first, then falls back to prometheus-k8s route. Use --metrics-backend to control which route is preferred.
[!WARNING]
kubeconfigauth mode requires a bearer token. Runoc whoami -tto verify you have one.If it fails, either:
- Re-login with:
oc login --token=<token>oroc login -u user -p password- Use port-forwarding with
--auth-mode headerinstead
Example using Prometheus as the preferred backend:
go run ./cmd/obs-mcp/ --listen 127.0.0.1:9100 --auth-mode kubeconfig --metrics-backend prometheus --insecure
Example using Thanos as the preferred backend:
[!NOTE]
Thanos versions before v0.40.0 do not expose the
/api/v1/status/tsdbendpoint, so guardrails that rely on TSDB stats (max-metric-cardinality,disallow-blanket-regexwithmax-label-cardinality > 0) will fail. Use--guardrails=noneor--guardrails='!tsdb'when using older Thanos versions. Thanos v0.40.0+ (#8484) added TSDB status support to the Query component, so guardrails should work if your cluster runs that version or later.
make run-no-guardrails
Or directly:
go run ./cmd/obs-mcp/ --listen 127.0.0.1:9100 --auth-mode kubeconfig --metrics-backend thanos --insecure --guardrails=none
[!IMPORTANT] How the Metrics Backend URL is Determined:
PROMETHEUS_URLenvironment variable (if set, always used)
--metrics-backendflag route discovery (only inkubeconfigmode)Default:
http://localhost:9090Example using explicit PROMETHEUS_URL:
PROMETHEUS_URL=https://thanos-querier.openshift-monitoring.svc:9091/ make run
[!IMPORTANT] How the Loki URL is determined (when
observability/logstoolset is enabled):
--loki-urlflagLOKI_URLenvironment variablelokiNamespace/lokiNametool parameters for each MCP tool call (exceptloki_list_instances)
[!IMPORTANT] How the Tempo URL is determined (when
observability/tracestoolset is enabled):
--traces.tempo-urlflagTEMPO_URLenvironment variabletempoNamespace/tempoNametool parameters for each MCP tool call (excepttempo_list_instances)
Port-forwards prometheus-k8s-0:9090 to localhost and starts obs-mcp with header auth. Requires oc login:
make run-openshift-pf-prometheus
Use the E2E test infrastructure for a fully working local environment with Prometheus:
make test-e2e-setup
This creates a Kind cluster with:
make test-e2e-deploy
kubectl port-forward -n obs-mcp svc/obs-mcp 9100:9100
To connect an MCP client, use http://localhost:9100/mcp.
When done:
make test-e2e-teardown
See TESTING.md for more details.
# add the prometheus-community Helm repo (once)
helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
helm repo update
# install Prometheus (and exporters) on your local cluster
helm install local prometheus-community/prometheus
# port-forward Prometheus server
export POD_NAME=$(kubectl get pods --namespace default -l "app.kubernetes.io/name=prometheus,app.kubernetes.io/instance=local" -o jsonpath="{.items[0].metadata.name}") && kubectl --namespace default port-forward $POD_NAME 9090
go run ./cmd/obs-mcp/ --auth-mode header --insecure --listen :9100
You can test the MCP server using curl. The server uses JSON-RPC 2.0 over HTTP.
[!TIP] For formatted JSON output, pipe the response to
jq:curl ... | jq
List available tools:
[!NOTE] The default
--toolsetsvalue isobservability/metricsonly. Additional toolsets:
observability/logs- Loki log query tools (requires Loki URL or LokiStack discovery)observability/traces- Tempo tracing tools (requires Tempo configuration)observability/otelcol- OpenTelemetry Collector configuration assistance (no external dependencies)Example:
--toolsets observability/metrics,observability/logs,observability/traces,observability/otelcol
curl -X POST http://localhost:9100/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'|jq
Call the list_metrics tool:
curl -X POST http://localhost:9100/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"list_metrics","arguments":{}}}' | jq
Execute a range query (e.g., get up metrics for the last hour):
curl -X POST http://localhost:9100/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"execute_range_query","arguments":{"query":"up{job=\"prometheus\"}","step":"1m","end":"NOW","duration":"1h"}}}' | jq
Use the MCP Inspector to visually test and debug obs-mcp tools.
Set up a Kind cluster with Prometheus and Alertmanager (if not already running):
make test-e2e-setup
Port-forward Prometheus and Alertmanager from your Kind cluster:
kubectl port-forward -n monitoring pod/prometheus-k8s-0 9090:9090 &
kubectl port-forward -n monitoring pod/alertmanager-main-0 9093:9093 &
Port-forward Prometheus and Alertmanager from your OpenShift cluster:
oc port-forward -n openshift-monitoring pod/prometheus-k8s-0 9090:9090 &
oc port-forward -n openshift-monitoring pod/alertmanager-main-0 9093:9093 &
Start obs-mcp and the Inspector (builds the obs-mcp container and starts both services via compose):
make inspect
This uses Docker by default. For podman, use:
CONTAINER_CLI=podman make inspect
Open the Inspector URL from the logs (includes the auth token):
http://localhost:6274/?MCP_PROXY_AUTH_TOKEN=<token>
Connect using Streamable HTTP transport to http://obs-mcp:8080/mcp
pkg/toolsets/{metrics,logs,traces,otelcol}; backend clients, discovery, authentication, and instrumentation live in upstream pkg/obs/.cmd/obs-mcp), Go MCP SDK adapter (pkg/mcp), health server, kubeconfig helpers, and standalone cluster-inspector stub. Only explicitly selected observability toolsets are exposed; standalone registration does not apply target-compatibility filtering.make generate-tools-doc here to regenerate TOOLS.md from the pinned upstream definitions; standalone integration and cluster e2e tests stay here.MCP server to allow LLMs to interact with a running Prometheus, Thanos, Alertmanager and optionally Tempo, OpenTelemetry Collector and Loki instances via the API
工具
尚無工具清單。遠端伺服器會在首次健康檢查後取得工具清單。
常見問題
在本頁複製適用於 Cursor、Claude Desktop、Claude Code 或 VS Code 的 stdio 設定。它使用發布者的安裝方式,而不是 PlugMCP 代理伺服器。
這筆記錄未列出必要的環境變數。
Obs 透過 stdio 在本機執行。
上次檢查時間:2026-11-13。 MCP server to allow LLMs to interact with a running Prometheus, Thanos, Alertmanager and optionally Tempo, OpenTelemetry Collector and Loki instances via the API