grafana

Build, modify, and ship Grafana dashboards as JSON via the HTTP API — panel types (timeseries / stat / gauge / table / heatmap / logs / traces / node-graph), gridPos 24-column layout, units, thresholds, template + datasource + chained variables, transformations (organize / calculateField / filterByValue), panel + dashboard links with ${field.labels.x} / ${from}, and Loki/Prometheus annotations. Use when scripting dashboard creation, writing the dashboard JSON for a new service, adding a $job dropdown variable, computing an "Error %" column with a transformation, overlaying deploys as annotations, or pushing a dashboard via POST /api/dashboards/db — even when the user says "create a dashboard for this metric", "add a service dropdown", "show errors as percentage", "overlay our deploys", or "export the dashboard JSON" without naming the API or schema. After every API push, verify with the returned version plus a GET on the dashboard UID.

Write, validate, and optimize PromQL for Prometheus / Grafana Mimir / Grafana Cloud Metrics. Covers rate vs irate vs increase, label matchers and regex, sum / avg / topk / by / without aggregation, classic + native histogramquantile, ratios with divide-by-zero guards, absent / changes for staleness, time offsets and predictlinear, recording-rule naming, SLO + burn-rate math, and a cardinality-hunting playbook. Use when writing a metric query, fixing wrong p95s, building an error-budget alert, debugging "query is slow", finding the noisy label that blew up cardinality, or migrating a dashboard query to a recording rule — even when the user says "calculate the error rate", "p99 latency", "sum by service", "why is this query slow", or "what's filling Mimir" without naming PromQL.

Grafana Loki log aggregation and LogQL query language. Covers LogQL syntax (log queries, metric queries, label matchers, line filters, parsers: json/logfmt/pattern/regexp/unpack, label filters, lineformat), Loki architecture, log ingestion via Alloy/Promtail/Fluent Bit, structured metadata, and Logs Drilldown. Use when writing LogQL queries, configuring Loki, troubleshooting log pipelines, or analyzing logs.

Expert evaluator for Grafana Loki label strategy. Audits, designs, and improves label schemas using cardinality scoring, access-pattern alignment, static vs. dynamic label rules, and consistency checks. Use when the user asks to evaluate, audit, design, or improve a Loki label strategy — or asks why their Loki queries are slow.

Instrument any app with OpenTelemetry and ship metrics / logs / traces to Grafana Cloud or self-hosted Mimir / Loki / Tempo / Pyroscope. Covers SDK auto-instrumentation for Go, Java (Grafana JVM agent), Python (opentelemetry-instrument), Node.js, .NET (Grafana.OpenTelemetry), Beyla eBPF for zero-code; Grafana Cloud OTLP gateway + Basic-auth (instanceID + API key, base64); env-var config (OTELEXPORTEROTLP, OTELRESOURCEATTRIBUTES); Alloy / OTel-Collector pipelines; Kubernetes Operator inject-annotations; and head + tail sampling. Use when instrumenting a service, pointing OTLP at Grafana Cloud, switching from Jaeger / Datadog / New Relic, choosing head- vs tail-sampling, or debugging "spans aren't showing in Explore" — even when the user says "auto-instrument my Java app", "send traces to Grafana", "what env vars do I set", "OTLP endpoint", or "Operator inject" without naming OpenTelemetry.

Configure Grafana OSS — provisions dashboards from YAML, sets up data sources (Prometheus / Loki / Tempo / Pyroscope), writes dashboard JSON with template variables, builds panel queries, assigns built-in roles (Viewer / Editor / Admin / GrafanaAdmin), mints service-account tokens, edits grafana.ini server config, creates annotations, installs plugins via provisioning, and validates each step with a health-check curl. Use when building dashboards, configuring data sources, setting up provisioning YAML, picking a panel type, writing template variables, managing users and roles, configuring SMTP/OAuth in grafana.ini, creating annotations via API, troubleshooting why a provisioned dashboard isn't showing up, or running Grafana OSS locally — even when the user says "set up a Prometheus data source", "provision dashboards from git", "make a service account", or "configure SSO in OSS" without saying "Grafana OSS".

Prometheus and Grafana Cloud Metrics overview including PromQL query language, Metrics Drilldown, alerting, recording rules, and integration patterns. Use when working with Prometheus, writing PromQL queries, configuring alerting, or discussing metrics architecture and best practices.

Configure Grafana Alerting, Incident Response Management (IRM), and SLOs end-to-end — provisions Grafana-managed and data-source-managed alert rules, contact points (Slack/PagerDuty/email/webhook), notification policies with hierarchical matchers, silences, mute timings, on-call schedules and escalation chains, incident-management integrations, and SLOs with multi-window burn-rate alerts. Use when configuring alerts, debugging notification routing, setting up on-call rotations, declaring or managing incidents, defining SLOs, provisioning alerting via YAML or API, picking matchers for a notification policy, building a PagerDuty/Slack webhook receiver, or troubleshooting why an alert isn't firing — even when the user says "page me on errors", "alert me when X happens", "route this to the platform team", or "set up an SLO" without naming Alerting or IRM.

Build a unified telemetry pipeline with Grafana Alloy — one OpenTelemetry-compatible binary that collects metrics, logs, traces, and profiles and ships to Grafana Cloud / Prometheus / Loki / Tempo / Pyroscope. Covers the Alloy config language (blocks, sys.env, component refs), prometheus.scrape → remotewrite, loki.source.file + loki.process → loki.write, otelcol.receiver.otlp → otelcol.exporter.otlp, pyroscope.scrape, K8s / Docker / EC2 discovery, relabeling, modules (import.file/git/http), clustering, Fleet Management remotecfg, the Alloy UI at :12345, and alloy fmt / alloy validate. Use when writing a config.alloy, replacing Grafana Agent / OTel Collector, scraping K8s pods, parsing logs, ingesting OTLP, or debugging "Alloy isn't sending anything" — even when the user says "set up the agent", "write me a scrape config", "drop these logs before sending", or "OTel collector config" without naming Alloy.

Connect AI coding agents (Claude Code, Cursor, VS Code, OpenAI Codex) to Grafana Cloud via the mcp-grafana Model Context Protocol server. Installs the server with go install, generates a Grafana service-account token, wires /.claude/settings.json or /.cursor/mcp.json with the command + env block, runs --disable-write for safer read-only sessions, switches to SSE transport for team-shared / VS Code setups, and verifies with /mcp + a listdatasources round-trip. Use when connecting Claude Code to Grafana, setting up MCP for Grafana, configuring the Grafana MCP server, using Grafana tools in Cursor/VS Code, querying Grafana from an AI agent, sharing the MCP server across a team — even when the user says "give my agent Grafana access", "let Claude see my metrics", or "Cursor + Grafana" without saying "MCP".

Stand up Grafana Tempo as a cost-efficient distributed-tracing backend that only needs object storage, and write TraceQL queries against it. Covers OTLP + Jaeger + Zipkin ingestion, the distributor → live-store → block-builder → object-storage write path, metrics-generator for RED spanmetrics + service graphs, Helm tempo-distributed deployment, multi-tenant X-Scope-OrgID, TraceQL span / resource / event scopes, structural operators (, <<), rate() + quantileovertime metrics, and the traces-to-logs / metrics / profiles datasource links. Use when deploying Tempo, writing a TraceQL query for slow / errored requests, debugging "no traces showing in Explore", sizing queriers / compactors, configuring S3 / GCS / Azure block storage, or wiring trace ↔ log ↔ profile correlation — even when the user says "tracing backend", "find slow requests", "show me the service graph", "store traces in S3", "Jaeger compatible store", or "what called this span" without naming Tempo.

Generate, validate, and review k6 test scripts — load, stress, spike, soak, smoke, breakpoint, functional, and protocol. Covers HTTP, WebSocket, gRPC, browser, all executors, thresholds, checks, custom metrics, the k6-testing library, k6 Cloud execution, and the xk6 extension ecosystem; uses the xk6-docs CLI (with grafana.com web fallback) for docs lookup and validates every script by running it. Use when writing, generating, validating, or debugging any k6 or load-test script (including plain-language asks like "load test this API" or "stress test my service"), choosing executors/scenarios, or setting thresholds. For end-to-end website performance suites use k6-perf-test-website; for documenting k6 itself use k6-docs; for a browser/UI test authored from a plain-language user journey (selectors discovered by exploring the live site) use k6-browser-test.

Auto-instrument an application's HTTP / gRPC / DB traffic with Grafana Beyla eBPF — no code changes, no SDK, no restart. Covers requirements (Linux 5.8+ with BTF, CAPSYSADMIN, host PID), language matrix (Go / Java / Python / Ruby / Node / .NET / Rust / C++ / PHP), Docker + Helm + DaemonSet install, port- / process- / Kubernetes-metadata discovery, OTLP traces + Prometheus metrics export, routes decorator (cardinality control), trace sampling, and Grafana Cloud via Alloy. Use when adding observability to a service you can't recompile, instrumenting a closed-source binary, getting RED metrics + spans onto Tempo/Mimir without touching the app, or rolling Beyla as a cluster-wide DaemonSet — even when the user says "zero-code APM", "instrument legacy app", "trace this binary", "eBPF observability", or "no SDK" without naming Beyla.

Stand up Grafana Mimir for horizontally scalable, multi-tenant, long-term Prometheus + OTLP metrics storage. Covers monolithic / read-write / microservices deployment, S3 / GCS / Azure / filesystem block storage, Prometheus remotewrite and OTLP ingestion, multi-tenancy with X-Scope-OrgID, ingester replication factor, compactor retention, and per-tenant limits. Use when running Mimir locally or on Kubernetes (Helm mimir-distributed), scaling Prometheus past a single node, picking ingest / query / backend split, configuring tenants and ingestion rate, debugging /ready 503s or 429 Too Many Requests, or pointing Grafana at a Mimir datasource — even when the user says "I need long-term Prometheus storage", "scale Prometheus", "multi-tenant metrics backend", "Cortex replacement", "remotewrite target", or "store 10M active series" without naming Mimir.

Continuously profile applications with Grafana Pyroscope and read the result as flame graphs. Covers three instrumentation paths — language SDK push (Go / Java / Python / Ruby / Node / .NET / Rust), Alloy eBPF auto-instrumentation (no code change, requires kernel 5.8+ with BTF), and SDK → Alloy receiver — plus ProfileQL queries, profile types (CPU / memory / allocations / goroutines / mutex), Grafana Cloud Profiles endpoint, and Span Profiles trace-to-profile linking. Use when adding profiling to a service, deploying Alloy as a cluster-wide eBPF profiler, hunting CPU / memory hotspots from a flame graph, comparing two profiles to find a regression, or correlating a slow Tempo trace to its profile — even when the user says "find what's burning CPU", "flame graph this app", "continuous profiling", "heap hotspots", or "why is allocation so high" without naming Pyroscope.

Get RED metrics + service maps + frontend RUM + AI/LLM monitoring out of Grafana Cloud — Application Observability (tracesspanmetrics from OTel traces, p50/p95/p99 latency, exemplar-to-trace, traces-to-logs / profiles), Frontend Observability with the Faro Web SDK (Core Web Vitals, session replay, pushError, React + router integration, TracingInstrumentation for browser → backend trace correlation), and AI Observability via OpenLIT (token / cost / latency, GPU, hallucination + toxicity evals). Use when standing up APM for a service, wiring an Alloy OTLP receiver + forwarding to Cloud, instrumenting a React frontend for RUM, debugging why service-map edges are missing, monitoring LLM cost drift, or correlating a frontend error to its backend trace — even when the user says "set up APM", "show service map", "monitor browser perf", "session replay", "RUM SDK", or "watch our OpenAI bill" without naming App / Frontend / AI Observability.

Ship Kubernetes, host, container, and cloud-provider telemetry into Grafana Cloud — k8s-monitoring Helm chart for K8s clusters (metrics + logs + traces + events + cost), Alloy prometheus.exporter.unix for Linux hosts, cAdvisor + Docker discovery for containers, and CloudWatch / Azure Monitor / Google Cloud Monitoring datasource setup. Use when onboarding a new cluster or VM fleet to Grafana Cloud, picking the right Helm values for K8s scraping, wiring kube-state-metrics + node-exporter + cAdvisor, alerting on PodCrashLooping / node memory / PVC capacity, or pulling AWS / Azure / GCP cloud metrics — even when the user says "monitor my cluster", "send K8s metrics to Grafana", "scrape EC2 metrics", "cluster pod logs", or "install the monitoring helm chart" without naming k8s-monitoring or Alloy.

Set up Grafana Cloud Database Observability for MySQL and PostgreSQL — enables pgstatstatements / Performance Schema, creates a least-privilege monitoring user, configures the databaseobservability.postgres / databaseobservability.mysql Alloy components, ships query samples + visual explain plans + RED metrics + schema details to Grafana Cloud, and correlates slow queries with application traces via db.statement / db.system OTel attributes. Use when monitoring database performance, diagnosing slow queries, setting up DB observability for RDS / Aurora / Cloud SQL / Azure Database / self-managed instances, correlating DB metrics with APM, or alerting on query latency — even when the user says "my database is slow", "find the slow queries", or "monitor RDS" without saying "observability".

Cut Grafana Cloud Metrics cost by shrinking active-series count with Adaptive Metrics aggregation rules — auto-recommendations from query history, custom exact/regex rules, label-drop config, unused-metric detection, and Alloy remotewrite fallback. Use when investigating a high Mimir/Grafana Cloud bill, hunting high-cardinality labels (poduid, serviceinstanceid, version), pre-aggregating counters/gauges, dropping unused metrics, or measuring grafanacloudinstanceactiveseries before/after — even when the user says "reduce cardinality", "too many series", "metrics spend", "active series count is exploding", or "drop the version label" without naming Adaptive Metrics.

Use when starting any grafana-app-sdk work — scaffolding a Grafana app, initializing a Grafana App Platform app, picking a deployment mode (standalone operator / grafana/apps / frontend-only), wiring app-specific config, or onboarding to the SDK. Covers grafana-app-sdk CLI install, project init per deployment mode, project layout, the schema-centric workflow (CUE kinds → generated code → reconciler/admission logic), and SpecificConfig for env-driven app configuration. Use even if the user just says "make a new app", "Grafana app platform", "kinds and watchers", "operator scaffold", or asks about apps/ inside the Grafana repo, without naming the SDK explicitly.

Implement reconcilers and watchers for grafana-app-sdk apps — write TypedReconciler[MyKind] reconcile functions, apply generation-based skip patterns, do conflict-safe status updates via resource.UpdateObject, configure BasicReconcileOptions (namespace, label/field filters, finalizer management), use Watcher for event-style handling, reconcile UnmanagedKinds (resources your app doesn't own), and register the whole thing in app.go. Use when writing a reconciler, implementing the reconcile loop, adding async business logic, handling create/update/delete events, processing resource state changes, scheduling periodic resyncs with RequeueAfter, picking between Watcher and Reconciler, or wiring a controller into app.go — even when the user says "process this resource", "handle X events", or "write a controller" without saying "reconciler".

Author CUE kind definitions for grafana-app-sdk apps - schemas, versioning, field constraints, named type definitions, custom routes, and codegen configuration. Scaffolds kinds via grafana-app-sdk project kind add, writes spec/status schemas with type constraints (regex, enum, range), defines -prefixed named types for reusable structs, registers versions in the app manifest, and runs grafana-app-sdk generate with explicit error recovery. Use when working with CUE kinds, adding a new resource type, adding a version to an existing kind, writing schema field constraints, defining Definition types, adding custom routes, editing files under kinds/, or when the user asks to "model a resource", "add a CUE schema", or "write a kind" — even without saying "CUE" explicitly.

Write or review k6 documentation across the three k6 repositories - k6-DefinitelyTyped (TypeScript types), k6-docs (user documentation), and k6 (release notes / changelog). Applies k6 doc style conventions, generates TypeScript type definitions, drafts release notes, and validates examples by running them against k6@master. Use when working on k6 documentation, the k6 changelog, the k6 release notes, k6 API reference, k6 TypeScript types, load testing docs, or when the user asks to write, edit, or review anything in the k6, k6-docs, or k6-DefinitelyTyped repos - even if they don't explicitly say "documentation".

Probe, load-test, and instrument frontends from Grafana Cloud. Covers Synthetic Monitoring (HTTP / DNS / TCP / Ping / Traceroute / Multihttp / k6-browser scripted checks from 20+ global probes, alert on probesuccess + TLS-cert expiry), Grafana Cloud k6 (distributed load tests across AWS load-zones, scenarios, httpreqduration thresholds, CI integration via grafana/k6-action), and Frontend Observability with Faro Web SDK (RUM, Core Web Vitals, custom events, pushError, distributed-trace correlation). Use when checking website / API uptime from multiple regions, gating a release on a load test, watching for TLS-cert renewal, instrumenting a React/Vue app, tracking Core Web Vitals, or correlating frontend errors to backend traces — even when the user says "is my login flow up?", "monitor my API", "ping our endpoint every minute", "release-gate load test", "browser performance monitoring", "session replay", or "RUM" without naming Synthetic Monitoring / k6 / Faro.

Cut your Grafana Cloud bill by attributing spend to teams and reducing telemetry volume. Covers FOCUS-compliant billing dashboards, cost-attribution labels in Alloy, Adaptive Metrics (cardinality reduction), Adaptive Logs (drop/sample), Adaptive Traces (tail sampling), usage alerts, and an optimization checklist. Use when investigating a high Grafana Cloud bill, attributing observability cost to a team or service, reducing active series / log bytes / trace spans, or setting up usage / quota alerts — even when the user says "our Grafana bill is too high", "who's burning the most metrics", "drop debug logs", "sample our traces", or "alert me before we hit quota" without naming Cost Management.

Manage a fleet of Grafana Alloy collectors with Fleet Management — author Alloy pipelines once, target them via attribute matchers (env="production", regex region="us-."), push remotely via OpAMP without restarting collectors. Covers pipeline create / update / matcher RPCs, collector attribute API, remotecfg bootstrap block (standalone + Helm), pre-deploy alloy fmt validation, the local Alloy UI at port 12345 for component health, and post-deploy REMOTECONFIGSTATUSAPPLIED verification. Use when standing up a Cloud Alloy fleet, pushing a config change to 200 collectors, hunting why one collector shows REMOTECONFIGSTATUSFAILED, validating River syntax before saving, or wiring discovery.kubernetes → prometheus.remotewrite — even when the user says "configure Alloy", "remote config the collectors", "push pipeline", "OpAMP", "collector is unhealthy", or "manage agent config centrally" without naming Fleet Management.

Sending telemetry data to Grafana Cloud — metrics via Prometheus remote write or OTLP, logs via Loki push or Alloy, traces via OTLP to Tempo, profiles via Pyroscope. Covers Alloy-based pipelines, direct SDK/agent integrations, cloud integrations catalog, and credentials management. Use when connecting an application or infrastructure to Grafana Cloud, setting up data ingestion, configuring remote write, or choosing between ingestion methods.

Route alerts, run on-call rotations, and drive incidents in Grafana IRM / OnCall — integrations (Alertmanager / Grafana Alerting / generic webhook / PagerDuty), Jinja2 routing + grouping templates, escalation chains (wait → notify schedule → notify team → webhook → auto-resolve), schedules (web + iCal + Terraform grafanaoncallschedule), Slack chatops with Acknowledge/Resolve/Silence, and the P1-P4 incident lifecycle. Use when wiring Alertmanager to OnCall, deciding which team gets paged, building rotations from a Google Calendar / iCal, hooking up Slack notifications, or declaring an incident from an alert — even when the user says "page the platform team on critical alerts", "send Prometheus alerts to Slack", "set up our on-call rota", "escalation policy", or "auto-resolve when the alert clears" without naming OnCall / IRM. Heads up — OnCall OSS is in maintenance mode (archived March 2026); Grafana Cloud users should use IRM.

Set up, configure, and troubleshoot Grafana Cloud integrations for AWS, Azure, and other cloud providers. Use when the user asks to connect AWS CloudWatch, set up Azure Monitor, configure Confluent Cloud observability, install a Grafana integration, set up hosted exporters, use AWS Firehose for CloudWatch logs, or troubleshoot a cloud integration. Triggers on phrases like "AWS CloudWatch", "Azure Monitor", "Confluent integration", "cloud integration", "hosted exporter", "AWS Firehose", "install integration", "cloud metrics", or "cloud logs".

Find the Prometheus metrics that drive your Grafana Cloud bill. dpm-finder is a Grafana Professional Services CLI that ranks metrics by Data Points per Minute (DPM) with per-label-set breakdown, optional --cost-per-1000-series pricing, and a Prometheus-exporter mode. Use when investigating high Grafana Cloud spend, hunting noisy / high-cardinality metrics, comparing pre/post recording-rule cardinality, or feeding cost data into dashboards — even when the user says "why is my Mimir bill so high?", "find the biggest metrics", "cardinality offenders", or "optimize Prometheus cost" without naming dpm-finder.

Build Grafana plugin pages using the @grafana/scenes framework. Use when creating new scene pages, adding panels/visualizations, setting up drilldown navigation, defining variables, configuring query runners, building table/timeseries/stat panels, or extending SceneObjectBase for custom scene objects. Triggers on any work involving SceneApp, SceneAppPage, EmbeddedScene, SceneQueryRunner, SceneDataTransformer, PanelBuilders, SceneFlexLayout, QueryVariable, or drilldown/tab configuration in Grafana plugins.

Use when the user asks to "write a validator", "add validation", "implement admission control", "write a mutating webhook", "add a mutation handler", "validate incoming resources", "implement admission logic", "add admission webhooks", "write ingress validation", or asks how to validate or mutate resources before they are persisted in a grafana-app-sdk app. Provides guidance on implementing validation and mutation admission handlers for grafana-app-sdk apps.

Optimise Grafana app plugin bundle size using React.lazy, Suspense, and webpack code splitting. Use when the user asks to reduce plugin bundle size, optimise module.js, add code splitting, improve initial plugin load performance, split plugin chunks, lazy load plugin pages, or help implement lazy loading in a Grafana app plugin. Triggers on phrases like "optimise plugin bundle size", "module.js is too large", "plugin is slow to load", "code split the plugin", "reduce initial JS payload", or "help me with Suspense in my plugin".

Manage Grafana Cloud accounts — organizations, stacks, RBAC roles and assignments, SSO/SAML/OAuth/GitHub auth, service accounts for CI/CD, user invites, team membership, and API-driven provisioning. Creates stacks via the Cloud API, mints service-account tokens, applies role assignments, configures SSO providers, and provisions teams/folders/dashboards via Terraform. Use when managing Grafana Cloud access, configuring SSO/SAML/OAuth, setting up service accounts for Terraform/CI/CD, assigning RBAC roles, inviting users, managing multiple stacks or organizations, provisioning cloud resources via API or Terraform, or auditing admin actions — even when the user says "set up SSO", "create a stack", "make a service account", or "onboard a team" without explicitly saying "admin".

Migrate a Grafana plugin to React 19 compatibility. Use when the user asks to update a plugin for React 19, prepare for React 19, fix React 19 compatibility, upgrade to React 19, migrate to React 19, bump grafanaDependency to 12.3.0, externalize jsx-runtime, or run react-detect. Triggers on phrases like "update plugin for React 19", "React 19 migration", "prepare for React 19", "plugin React 19 compat", "grafanaDependency 12.3.0", "JSX runtime externals", "react-detect", "SECRETINTERNALS", "ReactCurrentOwner", or "ReactCurrentDispatcher".

Turn on AI + ML features in Grafana Cloud — Grafana Assistant (NL → PromQL/LogQL/TraceQL, dashboard build, incident investigation, MCP integration), Dynamic Alerting (Prophet forecasting + DBSCAN outlier detection), Sift (8-analysis automated root-cause), Knowledge Graph + RCA Workbench, and the LLM Plugin (OpenAI / Anthropic / Azure / Ollama / vLLM / LiteLLM). Use when you want anomaly alerts without static thresholds, natural-language querying, automated incident investigation, dashboards generated from a sentence, or a managed LLM proxy for plugins — even when the user says "alert when something looks weird", "explain this PromQL", "find the root cause", "make this a dashboard", or "wire Claude into Grafana" without naming any of these products.

Set up private network connectivity to Grafana Cloud — AWS PrivateLink, Azure Private Link, GCP Private Service Connect, and Private Data Source Connect (PDC). Provisions VPC endpoints, private endpoints, or PSC forwarding rules per signal type (metrics / logs / traces / profiles); wires Alloy to push to the private DNS endpoint instead of the public one; verifies private DNS resolution + endpoint approval state. Use when sending telemetry to Grafana Cloud without traversing the public internet, eliminating cloud egress costs, meeting PCI-DSS / HIPAA / data-residency compliance, configuring PDC for private data-source queries, or connecting AWS / Azure / GCP workloads to Grafana — even when the user says "stop paying egress", "keep our telemetry off the public internet", or "PCI compliance for Grafana" without naming PrivateLink.

Expert evaluator for Prometheus label strategy on Grafana Cloud. Audits, designs, and improves label schemas using cardinality scoring, access-pattern alignment, static vs. dynamic label rules, histogram bucket discipline, and instrumentation hygiene. Prevents high cardinality at the source — in application code and scrape target labels — without dropping labels that make series unique (which breaks the data). For reducing the cost of series already in Grafana Cloud, routes to the adaptive-metrics skill. Use when the user asks to evaluate, audit, design, or improve Prometheus labels — or asks how to prevent high cardinality at the source. For "why is my Prometheus slow / expensive right now" triage, see prometheus-cardinality-troubleshooter.

Author, audit, and improve Grafana SKILL.md files against Anthropic's published Agent Skills guidance and the four-dimension rubric the grafana/skills CI gate uses (conciseness, actionability, workflow clarity, progressive disclosure). Applies the canonical SKILL.md structure (YAML frontmatter + body + references/ + scripts/ + assets/), the "pushy description" trigger pattern, the three-level progressive-disclosure model, and the validate-fix-rerun feedback loop. Use when creating a new skill in this repo, when reviewing a skill PR, when a skill's Tessl review score is below 75 (the merge gate), when a skill's description isn't getting picked up by agents, when restructuring a long SKILL.md into a bundle, or when the user asks how to write, improve, optimize, audit, or fix a skill - even if they don't say "skill" explicitly (e.g. "this isn't triggering", "Tessl scored this 72", "split this doc").

Clear description of what this skill does and when to use it. Use when the user asks about X or wants to work with Y. Include specific trigger phrases so agents auto-load it correctly. Max 1024 characters.

Diagnostic guide for active Prometheus cardinality problems — slow queries, OOMing Prometheus, high Grafana Cloud Active Series or DPM bills, "too many samples" ingest errors, series churn, or rapid memory growth. Walks through tsdb status endpoints, per-metric and per-label drill-downs, common-culprit galleries, and remediation paths. Use when the user is currently experiencing a cardinality fire. For preventing cardinality issues at the source, route to prometheus-label-strategy. For post-ingest aggregation, route to adaptive-metrics. For DPM-specific analysis, route to dpm-finder.

Use when the user wants to performance-test, load-test, or stress-test a public website end-to-end with k6. Produces a hybrid (protocol + browser) test suite, SLO-backed thresholds, a load-generator monitor sidecar, and a Grafana-side investigation playbook for backends the user owns. Triggers on "perf test my site", "performance test my site", "load test this URL", "stress test my web app", "I want to load test [URL]", "set up k6 against my website", "write a k6 suite for [site]", "see if my site handles N concurrent users", or "how does my site perform under traffic". Use this skill whenever the user mentions k6, load testing, stress testing, performance testing, or wants to validate a website under traffic — even if they don't explicitly use the word "test" or ask for the specific outputs this skill produces. This skill is for a full multi-test suite across several workflows; for a single browser test — functional OR a browser load test measuring Web Vitals for one journey — use k6-

Generate a copy-paste Grafana data source provisioning file (YAML or Terraform) for any plugin from its standardized settings schema on the plugins CDN. Use when the user wants to provision or configure a data source as code — e.g. "provision infinity", "datasource yaml for clickhouse", "terraform for the github datasource" — even when they only name the plugin and not the word "provisioning".

Interact with Grafana Cloud k6 (GCk6) — manage load tests, test runs, scripts, projects, schedules, env vars, fetch metrics or logs, and run scripts locally — using the gcx CLI (or direct curl when gcx is unavailable). Use this skill whenever the user mentions a k6 cloud test or run, asks to list/edit/create/start/abort k6 load tests, wants to fetch logs or metrics for a test run, manage k6 project limits or schedules, run a k6 script locally or via k6 cloud run, or needs to call any /cloud/v6/, /cloud/v5/, or k6-app Loki endpoint against a Grafana Cloud stack. Trigger even when the user doesn't explicitly say "gcx" or "API" — phrases like "why did my k6 test fail", "show me logs for run X", "bump VUh limit on project Y", or "update my k6 script" all qualify.

Reduces JavaScript dependency footprint with pnpm while preserving lockfile, workspace layout, and dependency range style. Runs /check-npm first, then removes unused deps, dedupes versions, ranks transitive closure, and reports Keep/Replace/Remove triage. Use when cleaning up pnpm dependencies, reducing lockfile size, or shrinking nodemodules in Grafana plugins; not for Go modules or full GitHub Actions workflow audits.

Audit a JavaScript/TypeScript repo's npm, yarn, or pnpm configuration for supply-chain hardening: tool version, lifecycle scripts, unsafe dependency protocols, and minimum release age ≥3 days. Use when the user invokes /check-npm or asks to audit package manager security, lifecycle scripts, git dependencies, ignore-scripts, min-release-age, allow-git, approvedGitRepositories, strictDepBuilds, or blockExoticSubdeps in a Grafana plugin or JS/TS project.

Maintain and improve existing k6 test scripts. Covers threshold tightening based on trend data, version migration between k6 releases, auto-fixing tests when the underlying service changes, refactoring for cleanliness, and auditing scripts against current best practices from docs. Use when the user asks to fix a failing k6 test, tighten thresholds, migrate a script to a new k6 version, refactor a test, update a script after a service change, or improve a script with best practices. Trigger on phrases like "fix my k6 test", "tighten my thresholds", "migrate to k6 v2", "update my test script", "refactor this k6 test", "my test is failing after a deploy", "apply best practices to my script", "modernize my k6 test", or "the service changed and my test broke". Also trigger when another skill (k6-trend-analysis or k6-cloud-investigate-test) hands off with a recommendation to edit a script. Maintains existing scripts; to author a new test use the k6 skill, or k6-browser-test for a browser tes

Analyze Grafana Cloud k6 test run trends over time. Detects slow metric drift (e.g., P95 latency creeping up while still passing thresholds), computes headroom to thresholds, flags anomalies, and recommends threshold tightening. Use when the user asks about test performance trends, wants to know if metrics are degrading, asks whether thresholds should be tightened, or wants a health check across recent runs for a specific test. Trigger on phrases like "how is my test trending", "is P95 getting worse", "check for performance regression", "should I tighten thresholds", "are my tests degrading", "show me trends for test X", "analyze my k6 test runs", or "is my test getting slower". Also trigger when a user asks to check all tests in a project -- run this skill once per test and synthesize.

Investigate a Grafana Cloud k6 test — describe the script, list run history, identify pass/fail status, pull raw metric time-series and log lines for one or more runs, and (if asked) safely edit the test script. Use when the user asks about a specific k6 cloud test or run, gives a /a/k6-app/tests/ or /a/k6-app/runs/ URL, asks "is this test passing", "why are my k6 tests failing", "show me metrics for run X", "show me logs for run X", or wants to add thresholds / fix a failing test. Trigger this skill even when the user doesn't explicitly say "investigate" — pasted /a/k6-app/... URLs, "why did run X fail", "what happened with my test", "is my test healthy", "the latest run looks weird", "checks are failing but the run says passed", or any request to diagnose a Grafana Cloud k6 run all qualify.

Author Grafana Cloud Synthetic Monitoring checks, with deep coverage of k6 scripted and browser checks: SM's single-VU/single-iteration execution model, assertions that actually fail probesuccess (expect() and fail() vs bare check()), secrets, deterministic scripts, robust browser locators, local validation with k6 run, deployment via UI/API/Terraform, verifying probesuccess, and rollback. Also helps choose the simplest sufficient check type (HTTP/ping/DNS/TCP, MultiHTTP, scripted, browser). Use when writing a synthetic check, monitoring a login/checkout/signup flow in production, converting a k6 script or an OpenAPI spec into a check, authoring a browser check, validating a user journey, or asking "is my site up from multiple regions". NOT for load, stress, or performance testing — SM runs one iteration per execution; for load tests use the grafana-k6 plugin or Grafana Cloud k6. For the broad Grafana Cloud Testing overview (SM + k6 Cloud + Faro), use the testing skill.