Rank #1 of 6 in Observability & Monitoring
Showcase


Try itExperimental
See what an agent can do with Grafana before you ever sign up. Pick a story: recorded sessions replay real probe-harness transcripts; sandboxed self-drive sessions are designed and gated (docs/TRY-IT.md).
$docker run -d --name pa-grafana-probe -p 13000:3000 grafana/grafana-oss && curl localhost:13000/api/health && curl -X POST http://admin:admin@localhost:13000/api/dashboards/db -d '{"dashboard":{"title":"PA Probe"},"overwrite":true}' && curl 'http://admin:admin@localhost:13000/api/search?query=PA%20Probe'recorded session — replayed, not liveVerified integrations
Connections to other tracked products — hover a chip for the verbatim evidence quote behind it.
By theme — the product's score on each story themeBy theme
Agenticness — how well agents can access and operate the productAgenticnessevidence →
How well agents can access and operate the product
Ai assist — stories about ai assist in this arenaAi assistevidence →
Stories about ai assist in this arena
Alerting slos — stories about alerting slos in this arenaAlerting slosevidence →
Stories about alerting slos in this arena
Automation depth — how much of the product can run unattendedAutomation depthevidence →
How much of the product can run unattended
Cost sampling — stories about cost sampling in this arenaCost samplingevidence →
Stories about cost sampling in this arena
Dashboards as code — stories about dashboards as code in this arenaDashboards as codeevidence →
Stories about dashboards as code in this arena
Deployment openness — stories about deployment openness in this arenaDeployment opennessevidence →
Stories about deployment openness in this arena
Incident response — stories about incident response in this arenaIncident responseevidence →
Stories about incident response in this arena
Openness — open source, data portability, and self-hosting storiesOpennessevidence →
Open source, data portability, and self-hosting stories
Otel standards — stories about otel standards in this arenaOtel standardsevidence →
Stories about otel standards in this arena
Privacy posture — data-handling and privacy storiesPrivacy postureevidence →
Data-handling and privacy stories
Query analytics — stories about query analytics in this arenaQuery analyticsevidence →
Stories about query analytics in this arena
Telemetry unified — stories about telemetry unified in this arenaTelemetry unifiedevidence →
Stories about telemetry unified in this arena
Story verdicts — every judged story with its evidenceStory verdicts
What’s free: 4 free · 1 paid · 0 enterprise · 33 not stated in evidence
Follow the green: where the map greys out is where Grafana stops today. ✓ full · ~ partial · ! disputed · — none · n/a not applicable.
Agenticness — how well agents can access and operate the productAgenticness
How well agents can access and operate the product
API surface
Drive the product through a documented public API
✓8/10
unlocks → Webhooks · Official SDKs · Scoped API keys · Versioning policy · API sandbox
Subscribe to events via webhooks
—–
Build against official SDKs
—0/10
Issue scoped/least-privilege API credentials for an agent
—0/10
Connect an agent via an official MCP server
✓9/10
Download a machine-readable API spec (OpenAPI or equivalent)
✓8/10
unlocks → Interactive API docs · Official SDKs
Rely on versioned APIs with a documented deprecation policy
—0/10
Test against a sandbox environment without touching production data
—0/10
Explore an interactive API reference with runnable examples
—0/10
Docs for agents
Point an agent at llms.txt or agent-oriented docs
✓9/10
Agentic features
Delegate tasks to a built-in AI assistant inside the product
~3/10
unlocks → MCP client
Operate the product with natural-language commands
~6/10
Plug MCP servers into this product so it can use their tools
—0/10
Get AI-generated insights and suggestions from my data inside the product
~5/10
Set up automations that run autonomously in the background
~5/10
Ai assist — stories about ai assist in this arenaAi assist
Stories about ai assist in this arena
Alerting slos — stories about alerting slos in this arenaAlerting slos
Stories about alerting slos in this arena
Automation depth — how much of the product can run unattendedAutomation depth
How much of the product can run unattended
Cost sampling — stories about cost sampling in this arenaCost sampling
Stories about cost sampling in this arena
Dashboards as code — stories about dashboards as code in this arenaDashboards as code
Stories about dashboards as code in this arena
Deployment openness — stories about deployment openness in this arenaDeployment openness
Stories about deployment openness in this arena
Incident response — stories about incident response in this arenaIncident response
Stories about incident response in this arena
Openness — open source, data portability, and self-hosting storiesOpenness
Open source, data portability, and self-hosting stories
Otel standards — stories about otel standards in this arenaOtel standards
Stories about otel standards in this arena
Privacy posture — data-handling and privacy storiesPrivacy posture
Data-handling and privacy stories
Query analytics — stories about query analytics in this arenaQuery analytics
Stories about query analytics in this arena
Telemetry unified — stories about telemetry unified in this arenaTelemetry unified
Stories about telemetry unified in this arena
Jump from a trace span to its correlated logs and metrics to debug a request end to end
✓8/10
Instrument hosts, containers, Kubernetes, and cloud services through vendor-maintained agents and integrations
~5/10
Collect metrics, logs, and traces in one platform and pivot between them with shared context
✓9/10
Sorted by importance (agentic first) (high → low) · 54/54 stories · click a row’s chevron for the rationale and evidence
Connect an agent via an official MCP server G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | full | 9/10 | Tprobed | |
Drive the product through a documented public API G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | full | 8/10 | Tprobed | |
Delegate tasks to a built-in AI assistant inside the product G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | partial | 3/10 | Tprobed | |
Plug MCP servers into this product so it can use their tools G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | none | 0/10 | ||
Point an agent at llms.txt or agent-oriented docs G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | full | 9/10 | Tprobed | |
Download a machine-readable API spec (OpenAPI or equivalent) G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | full | 8/10 | Tprobed | |
Run the product headlessly / in CI for automation G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | full | 8/10 | Tprobed | |
Operate the product with natural-language commands G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | partial | 6/10 | Tprobed | |
Use an official CLI G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | full | 6/10 | Tprobed | |
Get AI-generated insights and suggestions from my data inside the product G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | partial | 5/10 | Cclaimed | |
Set up automations that run autonomously in the background G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | partial | 5/10 | Cclaimed | |
Build against official SDKs G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | 0/10 | ||
Explore an interactive API reference with runnable examples G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | 0/10 | ||
Issue scoped/least-privilege API credentials for an agent G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | 0/10 | ||
Rely on versioned APIs with a documented deprecation policy G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | 0/10 | ||
Subscribe to events via webhooks G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | untested | none yet | |
Test against a sandbox environment without touching production data G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 1 | none | 0/10 | ||
Collect metrics, logs, and traces in one platform and pivot between them with shared context C Signals | sre | Telemetry unified — stories about telemetry unified in this arenaTelemetry unified | 3 | full | 9/10 | Xcommunity | |
Have an external agent query metrics, logs, and traces through documented APIs to debug production C Agent integration | ai-native user | Ai assist — stories about ai assist in this arenaAi assist | 3 | full | 9/10 | Tprobed | |
Alert on any telemetry signal with routing, grouping, and silencing of notifications C Alerting | sre | Alerting slos — stories about alerting slos in this arenaAlerting slos | 3 | full | 8/10 | Xcommunity | |
Analyze telemetry ad hoc with a documented query language C Query language | developer | Query analytics — stories about query analytics in this arenaQuery analytics | 3 | full | 8/10 | Xcommunity | |
Define dashboards and alerts as code (JSON models, Terraform, or API) and provision them repeatably C As code | developer | Dashboards as code — stories about dashboards as code in this arenaDashboards as code | 3 | full | 8/10 | Tprobed | |
Define rules that trigger actions automatically on events G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 3 | full | 7/10 | Xcommunity | |
Self-host the core product G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 3 | fullfree | 7/10 | Tprobed | |
Export all of my data in open formats and leave G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 3 | partial | 5/10 | Tprobed | |
Have the platform's AI investigate an alert or error and propose a probable root cause C Ai investigation | ai-native user | Ai assist — stories about ai assist in this arenaAi assist | 3 | partial | 4/10 | Tprobed | |
See what my observability spend is, attribute it to teams or services, and catch usage spikes before the bill C Cost | sre | Cost sampling — stories about cost sampling in this arenaCost sampling | 3 | partial | 4/10 | Xcommunity | |
Send telemetry directly over OTLP with first-class OpenTelemetry support C Otel | developer | Otel standards — stories about otel standards in this arenaOtel standards | 3 | none | 0/10 | ||
Prevent my data from being used to train AI models G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 3 | n/a | untested | none yet | |
Build shareable dashboards with rich visualization types and template variables C Dashboards | sre | Dashboards as code — stories about dashboards as code in this arenaDashboards as code | 2 | full | 8/10 | Xcommunity | |
Declare and track incidents with timelines, on-call schedules, and escalation policies C Incidents | sre | Incident response — stories about incident response in this arenaIncident response | 2 | fullpaid | 8/10 | Xcommunity | |
Define SLOs with error budgets and burn-rate alerts C Slos | sre | Alerting slos — stories about alerting slos in this arenaAlerting slos | 2 | full | 8/10 | Cclaimed | |
Jump from a trace span to its correlated logs and metrics to debug a request end to end C Correlation | developer | Telemetry unified — stories about telemetry unified in this arenaTelemetry unified | 2 | full | 8/10 | Cclaimed | |
Read the product's source under an open license G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 2 | fullfree | 8/10 | Xcommunity | |
Run the full observability stack self-hosted in production with documented architecture and upgrade path G Self host | sre | Deployment openness — stories about deployment openness in this arenaDeployment openness | 2 | partialfree | 7/10 | Tprobed | |
Do everything through the API that I can do in the UI G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 2 | partial | 6/10 | Tprobed | |
Ask questions of my telemetry in natural language and get a real query or chart back C Ai querying | ai-native user | Ai assist — stories about ai assist in this arenaAi assist | 2 | partial | 5/10 | Tprobed | |
Control trace/log sampling and retention tiers to manage data volume deliberately C Sampling | developer | Cost sampling — stories about cost sampling in this arenaCost sampling | 2 | partial | 5/10 | Xcommunity | |
Group and filter by high-cardinality fields (user id, request id) without pre-aggregating or defining indexes first C Analysis | developer | Query analytics — stories about query analytics in this arenaQuery analytics | 2 | partial | 5/10 | Cclaimed | |
Instrument hosts, containers, Kubernetes, and cloud services through vendor-maintained agents and integrations C Instrumentation | sre | Telemetry unified — stories about telemetry unified in this arenaTelemetry unified | 2 | partial | 5/10 | Xcommunity | |
Instrument once with open standards and switch backends without re-instrumenting my code C Otel | sre | Otel standards — stories about otel standards in this arenaOtel standards | 2 | partial | 5/10 | Xcommunity | |
Perform bulk operations across many items at once G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 2 | partial | 5/10 | Tprobed | |
Correlate regressions with deploys and configuration changes via release or change tracking C Change tracking | developer | Incident response — stories about incident response in this arenaIncident response | 2 | partial | 3/10 | Cclaimed | |
Get AI-generated summaries of incidents and alert context for responders C Ai investigation | ai-native user | Ai assist — stories about ai assist in this arenaAi assist | 2 | partial | 3/10 | Cclaimed | |
Schedule recurring jobs or workflows G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 2 | none | 0/10 | ||
Choose where my data is stored (region/residency) G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 2 | none | untested | none yet | |
Control data retention and deletion G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 2 | none | untested | none yet | |
Opt out of telemetry and usage tracking G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 2 | none | untested | none yet | |
Point alert notifications at webhooks that trigger automated remediation or agents G Alert automation | ai-native user | Alerting slos — stories about alerting slos in this arenaAlerting slos | 2 | none | untested | none yet | |
See application errors grouped into issues with stack traces, release tracking, and regression detection C Errors | developer | Query analytics — stories about query analytics in this arenaQuery analytics | 2 | none | untested | none yet | |
Spin up a local or dev instance of the platform to test instrumentation and dashboards C Local dev | developer | Deployment openness — stories about deployment openness in this arenaDeployment openness | 1 | partial | 6/10 | Tprobed | |
Version, review, and roll back my automations G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 1 | partial | 6/10 | Cclaimed | |
Predict costs from transparent published per-signal pricing without talking to sales G Cost | sre | Cost sampling — stories about cost sampling in this arenaCost sampling | 1 | partialfree | 5/10 | Xcommunity | |
Enable anomaly or outlier detection that surfaces problems without hand-written thresholds C Alerting | sre | Alerting slos — stories about alerting slos in this arenaAlerting slos | 1 | none | untested | none yet |
Opportunities — the stories that would move this product's scores, from its own judged verdictsOpportunitiestop 8 of 35 stories with headroom
What would move Grafana’s scores — derived from its own judged verdicts, biggest headroom first. Each line quotes what the judge found missing; shipping it (or evidencing it publicly) is the fix.
Agenticness — how well agents can access and operate the productPlug MCP servers into this product so it can use their tools
nonemoves agent-readyimpact 45
The evidence shows Grafana ships an MCP *server* (Cloud MCP Server / self-managed grafana-mcp) that lets external AI agents call Grafana's own tools — this is the reverse direction of the story, which asks whether Grafana itself can plug in and consume external MCP servers' tools.
Agenticness — how well agents can access and operate the productDelegate tasks to a built-in AI assistant inside the product
partialq3/10moves Built-in AIimpact 31.5
Missing: description of an actual built-in assistant interface, concrete examples of delegated tasks/actions it performs, and independent/community confirmation it works.
Otel standards — stories about otel standards in this arenaSend telemetry directly over OTLP with first-class OpenTelemetry support
nonemoves PA Scoreimpact 30
The evidence pack never mentions OTLP, OpenTelemetry SDKs, or native OTLP ingestion endpoints in Grafana; the closest reference is sending logs to Loki via Grafana Alloy, which is an indirect, unrelated mention rather than documentation of first-class OTLP support in Grafana itself.
Agenticness — how well agents can access and operate the productIssue scoped/least-privilege API credentials for an agent
nonemoves agent-readyimpact 30
The evidence pack shows Grafana has an MCP server for agent connectivity (grafana-docs-7, grafana-probe-4) but contains no mention of scoped or least-privilege API keys, service accounts, or role-based credential issuance for agents.
Agenticness — how well agents can access and operate the productBuild against official SDKs
nonemoves agent-readyimpact 30
Missing: documented official SDK packages/libraries, SDK usage examples, or client-library release notes.
Agenticness — how well agents can access and operate the productSubscribe to events via webhooks
nonemoves agent-readyimpact 30
Missing: any citation describing webhook contact points/notifiers or an event subscription API.
Agenticness — how well agents can access and operate the productExplore an interactive API reference with runnable examples
nonemoves API qualityimpact 30
Evidence shows Grafana exposes an OpenAPI spec (grafana-probe-3) but nothing indicates an interactive reference UI with runnable/try-it-out examples; docs excerpts focus on dashboards, alerting, and data sources, not API exploration tooling.
Agenticness — how well agents can access and operate the productRely on versioned APIs with a documented deprecation policy
nonemoves API qualityimpact 30
There's an OpenAPI spec probe confirming an API exists, but no evidence of API versioning scheme or a documented deprecation policy for that API; no changelog/deprecation policy citations appear anywhere in the pack.
Showing the top 8 of 35 — every none/partial verdict in the story verdicts table is headroom.
Think a verdict is wrong? Every verdicts-table row has a Flag link — see the methodology.
Coverage map — which docs area, API section, or community source covers which judged storiesCoverage map8 surfaces · 38 covered stories
Where the cited evidence behind each covered verdict came from — the same citations the verdicts table shows, no extra judging.
docs33 stories
- Point an agent at llms.txt or agent-oriented docs
- Run the product headlessly / in CI for automation
- Connect an agent via an official MCP server
- Use an official CLI
- Drive the product through a documented public API
- Get AI-generated insights and suggestions from my data inside the product
- Set up automations that run autonomously in the background
- Delegate tasks to a built-in AI assistant inside the product
- Operate the product with natural-language commands
- Have an external agent query metrics, logs, and traces through documented APIs to debug production
- Get AI-generated summaries of incidents and alert context for responders
- Have the platform's AI investigate an alert or error and propose a probable root cause
- Ask questions of my telemetry in natural language and get a real query or chart back
- Alert on any telemetry signal with routing, grouping, and silencing of notifications
- Define SLOs with error budgets and burn-rate alerts
- Perform bulk operations across many items at once
- Define rules that trigger actions automatically on events
- Version, review, and roll back my automations
- Define dashboards and alerts as code (JSON models, Terraform, or API) and provision them repeatably
- Build shareable dashboards with rich visualization types and template variables
- Spin up a local or dev instance of the platform to test instrumentation and dashboards
- Run the full observability stack self-hosted in production with documented architecture and upgrade path
- Correlate regressions with deploys and configuration changes via release or change tracking
- Declare and track incidents with timelines, on-call schedules, and escalation policies
- Do everything through the API that I can do in the UI
- Export all of my data in open formats and leave
- Self-host the core product
- Instrument once with open standards and switch backends without re-instrumenting my code
- Group and filter by high-cardinality fields (user id, request id) without pre-aggregating or defining indexes first
- Analyze telemetry ad hoc with a documented query language
- Jump from a trace span to its correlated logs and metrics to debug a request end to end
- Instrument hosts, containers, Kubernetes, and cloud services through vendor-maintained agents and integrations
- Collect metrics, logs, and traces in one platform and pivot between them with shared context
GitHub README15 stories
- Connect an agent via an official MCP server
- Drive the product through a documented public API
- Delegate tasks to a built-in AI assistant inside the product
- Operate the product with natural-language commands
- Have an external agent query metrics, logs, and traces through documented APIs to debug production
- Have the platform's AI investigate an alert or error and propose a probable root cause
- Ask questions of my telemetry in natural language and get a real query or chart back
- Build shareable dashboards with rich visualization types and template variables
- Do everything through the API that I can do in the UI
- Read the product's source under an open license
- Instrument once with open standards and switch backends without re-instrumenting my code
- Group and filter by high-cardinality fields (user id, request id) without pre-aggregating or defining indexes first
- Analyze telemetry ad hoc with a documented query language
- Jump from a trace span to its correlated logs and metrics to debug a request end to end
- Collect metrics, logs, and traces in one platform and pivot between them with shared context
Hacker News15 stories
- Alert on any telemetry signal with routing, grouping, and silencing of notifications
- Define rules that trigger actions automatically on events
- See what my observability spend is, attribute it to teams or services, and catch usage spikes before the bill
- Predict costs from transparent published per-signal pricing without talking to sales
- Control trace/log sampling and retention tiers to manage data volume deliberately
- Build shareable dashboards with rich visualization types and template variables
- Run the full observability stack self-hosted in production with documented architecture and upgrade path
- Declare and track incidents with timelines, on-call schedules, and escalation policies
- Do everything through the API that I can do in the UI
- Read the product's source under an open license
- Self-host the core product
- Instrument once with open standards and switch backends without re-instrumenting my code
- Analyze telemetry ad hoc with a documented query language
- Instrument hosts, containers, Kubernetes, and cloud services through vendor-maintained agents and integrations
- Collect metrics, logs, and traces in one platform and pivot between them with shared context
Products docs10 stories
- Connect an agent via an official MCP server
- Get AI-generated insights and suggestions from my data inside the product
- Delegate tasks to a built-in AI assistant inside the product
- Operate the product with natural-language commands
- Have an external agent query metrics, logs, and traces through documented APIs to debug production
- Get AI-generated summaries of incidents and alert context for responders
- Have the platform's AI investigate an alert or error and propose a probable root cause
- Ask questions of my telemetry in natural language and get a real query or chart back
- See what my observability spend is, attribute it to teams or services, and catch usage spikes before the bill
- Control trace/log sampling and retention tiers to manage data volume deliberately
API reference7 stories
- Run the product headlessly / in CI for automation
- Drive the product through a documented public API
- Download a machine-readable API spec (OpenAPI or equivalent)
- Have an external agent query metrics, logs, and traces through documented APIs to debug production
- Perform bulk operations across many items at once
- Define dashboards and alerts as code (JSON models, Terraform, or API) and provision them repeatably
- Do everything through the API that I can do in the UI
grafana.com3 stories
llms.txt2 stories
Probe proofs — replayable recordings from the probe harnessProbe proofs
Replayable recordings from our probe harness — see the Prove-It protocol to submit one.
$docker run -d --name pa-grafana-probe -p 13000:3000 grafana/grafana-oss && curl localhost:13000/api/health && curl -X POST http://admin:admin@localhost:13000/api/dashboards/db -d '{"dashboard":{"title":"PA Probe"},"overwrite":true}' && curl 'http://admin:admin@localhost:13000/api/search?query=PA%20Probe'reproduced$ docker run -d --name pa-grafana-probe -p 13000:3000 grafana/grafana-oss && curl localhost:13000/api/health && curl -X POST http://admin:admin@localhost:13000/api/dashboards/db -d '{"dashboard":{"title":"PA Probe"},"overwrite":true}' && curl 'http://admin:admin@localhost:13000/api/search?query=PA%20Probe'
{
"database": "ok",
"version": "13.0.2",
"commit": "3fcdbc5a"
}
{"folderUid":"","id":1869679639199744,"slug":"pa-probe","status":"success","uid":"8059113c-7db4-4e26-bba3-0389c84f569e","url":"/d/8059113c-7db4-4e26-bba3-0389c84f569e/pa-probe","version":1}
[{"id":1869679639199744,"uid":"8059113c-7db4-4e26-bba3-0389c84f569e","orgId":1,"title":"PA Probe","uri":"db/pa-probe","url":"/d/8059113c-7db4-4e26-bba3-0389c84f569e/pa-probe","slug":"","type":"dash-db","tags":[],"isStarred":false,"sortMeta":0,"isDeleted":false}]
$echo '<jsonrpc initialize>' | mcp-grafanareproduced$ echo '<jsonrpc initialize>' | mcp-grafana
time=2026-09-04T18:11:20.973-07:00 level=INFO msg="Using Grafana configuration" url=http://localhost:3000 api_[redacted]_set=false basic_auth_set=false org_id=0 extra_headers_count=0
time=2026-09-04T18:11:20.974-07:00 level=WARN msg="Failed to fetch frontend settings" error="Get \"http://localhost:3000/api/frontend/settings\": dial tcp [::1]:3000: connect: connection refused"
time=2026-09-04T18:11:20.975-07:00 level=ERROR msg="failed to initialize proxied tools for stdio" error="failed to discover MCP datasources: failed to list datasources: Get \"http://localhost:3000/api/datasources\": dial tcp [::1]:3000: connect: connection refused"
time=2026-09-04T18:11:20.975-07:00 level=INFO msg="Starting Grafana MCP server using stdio transport" version=(devel)
time=2026-09-04T18:11:20.975-07:00 level=INFO msg="Using Grafana configuration" url=http://localhost:3000 api_[redacted]_set=false basic_auth_set=false org_id=0 extra_headers_count=0
time=2026-09-04T18:11:20.975-07:00 level=WARN msg="Failed to fetch frontend settings" error="Get \"http://localhost:3000/api/frontend/settings\": dial tcp [::1]:3000: connect: connection refused"
{"jsonrpc":"2.0","id":1,"result":{"protocolVersion":"2025-06-18","capabilities":{"resources":{},"tools":{"listChanged":true}},"serverInfo":{"name":"mcp-grafana","version":"(devel)"},"instructions":"This server provides access to your Grafana instance and the surrounding ecosystem.\n\nAvailable Capabilities:\n- Search: Find dashboards, folders, and other Grafana resources.\n- Datasources: List and fetch details for datasources.\n- Incidents: Search, create, update, and resolve incidents in Grafana Incident.\n- Prometheus: Run PromQL queries, retrieve metric metadata, and explore label names/values.\n- Loki: Run LogQL queries, retrieve log metadata, and explore label names/values.\n- Alerting: List and fetch alert rules and notification contact points.\n- Dashboards: Search, retrieve, update, and create dashboards. Extract panel queries and datasource information.\n- Folders: Manage dashboard folders.\n- OnCall: View and manage on-call schedules, shifts, teams, and users.\n- Asserts: Query and analyze assertion data.\n- Sift Investigations: Start and manage Sift investigations, analyze logs/traces, find error patterns, and detect slow requests.\n- Pyroscope: Profile applications and fetch profiling data.\n- Navigation: Generate deeplink URLs for Grafana resources like dashboards, panels, and Explore queries, with optional built-in shortening.\n- Annotations: Create and manage dashboard annotations.\n- Rendering: Export dashboard panels or full dashboards as PNG images (requires Grafana Image Renderer plugin).\n- Snapshots: List, get, create, and delete dashboard snapshots.\n- Plugins: Check whether Grafana plugins are installed and fetch plugin details.\n- API: Make authenticated HTTP requests to any Grafana API endpoint with optional jq-style response filtering.\n- Config: Generate operator-facing configuration snippets (e.g. Alloy label-enforcement pipelines).\n- Provisioning: List provisioning repositories (e.g. git-sync sources) to discover repository slugs for use with rendering tools.\n- Docs: Search and retrieve Grafana product documentation (powered by grafana.com/llms-full.txt).\n- User: Identify the current user/credential, its capabilities, and the organizations it can access.\n- Proxied Tools: Access tools from external MCP servers (like Tempo) through dynamic discovery.\n\nTimestamp parameters without a timezone offset are interpreted as UTC. Include an offset like '-05:00' or use relative syntax like 'now-1h' to query in a different timezone.\n"}}
Claims vs evidence — vendor claims reconciled against independent verdictsClaims vs evidence
13 of 15 testable claims verified · 0 contradicted → integrity 87/100
23 distinct capability claims found in Grafana’s own claimed-docs/GitHub materials, reconciled against our judge’s independent verdicts.
13
Verified
2
Unverified
0
Contradicted
23
Undersold
Verified (19)
“Query and combine data across multiple data sources regardless of where data is stored”
Collect metrics, logs, and traces in one platform and pivot between them with shared contextfullproof ↗
“Create, manage, and act on all alerts from a single consolidated view”
Alert on any telemetry signal with routing, grouping, and silencing of notificationsfullproof ↗
“Provision dashboards and data sources from version-controlled config files (GitOps)”
Define dashboards and alerts as code (JSON models, Terraform, or API) and provision them repeatablyfullproof ↗
“Define dashboards, data sources, and configs as code with version control, testing, and CI/CD deployment”
Define dashboards and alerts as code (JSON models, Terraform, or API) and provision them repeatablyfullproof ↗
“Create on-call schedules, escalate alerts, and coordinate incidents through post-incident review”
Declare and track incidents with timelines, on-call schedules, and escalation policiesfullproof ↗
“AI agents can connect to Grafana via a hosted Cloud MCP Server or self-managed open-source MCP Server”
“Understand production systems faster using natural language queries”
Ask questions of my telemetry in natural language and get a real query or chart backpartialproof ↗
“Explore data with ad-hoc queries and dynamic drilldown, splitting/comparing time ranges and sources side by side”
Build shareable dashboards with rich visualization types and template variablesfullproof ↗
“Create dynamic, reusable dashboards using dropdown template variables”
Build shareable dashboards with rich visualization types and template variablesfullproof ↗
“Mix different data sources, including custom ones, within the same graph on a per-query basis”
Collect metrics, logs, and traces in one platform and pivot between them with shared contextfullproof ↗
“Unified platform to query, visualize, alert on, and explore metrics, logs, and traces wherever stored”
Collect metrics, logs, and traces in one platform and pivot between them with shared contextfullproof ↗
“Free tier plan available with no charges for personal projects and startups”
Predict costs from transparent published per-signal pricing without talking to salespartialproof ↗
“Adaptive Telemetry automatically identifies important data and aggregates the rest, cutting telemetry costs up to 80%”
See what my observability spend is, attribute it to teams or services, and catch usage spikes before the billpartialproof ↗
“LogQL query language uses labels and operators to filter and analyze logs in Loki”
Analyze telemetry ad hoc with a documented query languagefullproof ↗
“Documented migration and upgrade paths for existing Loki deployments”
Run the full observability stack self-hosted in production with documented architecture and upgrade pathpartialproof ↗
“Grafana Alerting notifies you about system problems moments after they occur”
Alert on any telemetry signal with routing, grouping, and silencing of notificationsfullproof ↗
“Official CLI supports authentication, managing multiple environments, and admin tasks, suited for CI/CD and local dev”
“During an incident, investigate using the same metrics, logs, and traces that triggered the alert”
Declare and track incidents with timelines, on-call schedules, and escalation policiesfullproof ↗
“Run Loki locally in single-binary mode with a local filesystem backend, sending logs via Grafana Alloy”
Spin up a local or dev instance of the platform to test instrumentation and dashboardspartialproof ↗
Unverified (4)
“Tempo lets you search traces, generate metrics from spans, and link traces with logs and metrics”
Jump from a trace span to its correlated logs and metrics to debug a request end to endfullproof ↗
“Grafana SLO framework lets you define SLIs/SLOs to measure and act on service quality”
Define SLOs with error budgets and burn-rate alertsfullproof ↗
“Switch seamlessly from metrics to logs with preserved label filters, including live log streaming”
Jump from a trace span to its correlated logs and metrics to debug a request end to endfullproof ↗
“Prometheus exemplars let you jump directly from metrics to related Tempo traces”
Jump from a trace span to its correlated logs and metrics to debug a request end to endfullproof ↗
Undersold (23)
Point an agent at llms.txt or agent-oriented docsfullproof ↗
Run the product headlessly / in CI for automationfullproof ↗
Drive the product through a documented public APIfullproof ↗
Get AI-generated insights and suggestions from my data inside the productpartialproof ↗
Set up automations that run autonomously in the backgroundpartialproof ↗
Delegate tasks to a built-in AI assistant inside the productpartialproof ↗
Operate the product with natural-language commandspartialproof ↗
Download a machine-readable API spec (OpenAPI or equivalent)fullproof ↗
Have an external agent query metrics, logs, and traces through documented APIs to debug productionfullproof ↗
Get AI-generated summaries of incidents and alert context for responderspartialproof ↗
Have the platform's AI investigate an alert or error and propose a probable root causepartialproof ↗
Perform bulk operations across many items at oncepartialproof ↗
Define rules that trigger actions automatically on eventsfullproof ↗
Control trace/log sampling and retention tiers to manage data volume deliberatelypartialproof ↗
Correlate regressions with deploys and configuration changes via release or change trackingpartialproof ↗
Do everything through the API that I can do in the UIpartialproof ↗
Export all of my data in open formats and leavepartialproof ↗
Instrument once with open standards and switch backends without re-instrumenting my codepartialproof ↗
Group and filter by high-cardinality fields (user id, request id) without pre-aggregating or defining indexes firstpartialproof ↗
Instrument hosts, containers, Kubernetes, and cloud services through vendor-maintained agents and integrationspartialproof ↗
Business model
AGPL open-source Grafana/Loki/Tempo/Mimir are free to self-host; Grafana Cloud has a generous free tier then usage-based pricing per metrics/logs/traces volume, plus enterprise contracts.
pricing ↗Score trend
How this product’s scores have moved as evidence and verdicts are re-derived — a point per change, not per day.
Try Experimental
Run it in the microterminal →Recorded agent sessions — and a live MCP handshake where the vendor ships one.
Flag
⚑ Flag a verdictThink a verdict is wrong? Opens a prefilled GitHub issue — or use the ⚑ next to any verdict above.
For agents
Agent surface uptime llms.txt 100% · openapi.json 100% (30d, checked every 6h since Sep 8 '26)
