{ "data": { "a": { "slug": "better-stack", "name": "Better Stack", "vendor": "Better Stack, Inc.", "vendorUrl": "https://betterstack.com", "kind": "mcp", "category": "observability", "summary": "Better Stack is a hosted observability and incident management service from Better Stack, Inc. with uptime monitors, on-call, status pages, logs, metrics, traces and error tracking, reached through REST APIs and a remote MCP server.", "url": "https://www.anchorterminal.com/tools/better-stack", "markdownUrl": "https://www.anchorterminal.com/tools/better-stack.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/better-stack.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/better-stack.json", "dns": "better-stack.grade.anchorterminal.com", "repo": "https://github.com/BetterStackHQ/claude-plugin", "license": "Proprietary hosted service under the Better Stack Terms of Use. The Claude, Cursor and Codex plugins and the Terraform providers on GitHub are MIT", "transports": [ "streamable-http" ], "remoteUrl": "https://mcp.betterstack.com", "packages": [], "auth": "mixed", "authNotes": "OAuth sign-in in the browser is the documented default for the MCP server, with a Full access level chosen at consent for writes. Unattended agents send an API token as a Bearer header. Tokens are minted in the web app under API tokens, either global (every team) or team-based (Incidents API or Telemetry API tokens), with Read \u0026 write or read access per the MCP docs. No scopes finer than that and no key expiry were found. Access is self-serve after a browser signup.", "pricing": "freemium", "pricingNotes": "The MCP server and APIs carry no separate charge. The Free plan has 10 monitors and heartbeats, one status page, 3 GB of logs and traces kept 3 days, 30 GB of metrics and 100,000 errors a month, with no card mentioned on the signup page. Responders cost $34 a month ($29 yearly), telemetry is priced per GB by region ($0.10 ingestion and $0.05 retention per GB in Europe, $0.15 and $0.08 in the US, $0.35 and $0.18 in Singapore) or in bundles from $30 a month, and audit logs are a $250 a month add-on (https://betterstack.com/pricing, checked 2026-11-03).", "priceSummary": "$34 / seat-mo", "where": "hosted", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the pricing page, the API docs or the MCP docs (checked 2026-11-03). Plans are paid by card through the web app.", "endpoints": [] }, "toolCount": 148, "popularity": { "githubStars": 6, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-11-03" }, "docsUrl": "https://betterstack.com/docs/getting-started/integrations/mcp/", "capabilities": [ "observability.logs", "observability.incidents", "work.oncall", "observability.metrics", "observability.traces", "observability.errors" ], "tags": [ "official", "hosted", "closed-source", "mcp", "remote-mcp", "oauth", "api-key", "toolsets", "free-tier", "freemium", "eu-hosted", "status-page", "audit-log", "soc2", "webhooks" ], "lastRelease": "2026-11-02", "graded": true, "anchor": { "graded": true, "score": 54.2, "grade": "C", "agentReady": false, "rank": 811, "ranked": true, "rankOf": 1349, "categoryRank": 7, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 55, "maintenance": 71, "payments": 40, "reliability": 50, "schema": 46, "security": 60, "transparency": 69 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-03" }, "negative": 0, "verdict": "One remote MCP server reaches incidents, on-call, monitors, status pages, logs, metrics, traces and errors with OAuth or a team-scoped token, and two headers cut its 148 tools down. Tool schemas and rate limits aren't published, there is no SLA, and audit logs of token use are a paid add-on.", "bestFor": "An agent that investigates incidents, queries logs and metrics with SQL, manages monitors and posts status page updates for a team already on Better Stack, with the Claude plugin as the shortest route.", "strengths": [ "148 tools across incidents, on-call, monitors, heartbeats, status pages, catalogue, telemetry queries, dashboards, alerts, errors and team members, on one remote endpoint", "X-MCP-Tools-Only and X-MCP-Tools-Except headers give a read-only or task-sized subset, and OAuth consent or a token can be read-level", "Audit logs record API token requests with the token's name and the actor type, and can be streamed to a source or any URL", "Free plan with 10 monitors, a status page and 3 GB of logs, and per-GB prices published by region", "Dated changelog entries on 10 August, 9 September, 1 October, 15 October and 29 October 2026, and a plugin release on 2 November 2026" ], "weaknesses": [ "No rate limit numbers in the API docs, and the only 429 guidance is to wait before retrying", "The MCP tool schemas are not published, the REST API has no OpenAPI file and the docs have no llms.txt", "No SLA. The terms supply the service as is, and the enterprise page lists SLA reporting as a feature", "Delayed Telemetry data processing in the Europe region ran from 3 to 6 September 2026 in entries of 6 hours to 1 day", "Audit logs cost $250 a month, and no security.txt, bug bounty or disclosure policy was found" ], "agentNotes": [ "Connect with `claude mcp add betterstack --transport http https://mcp.betterstack.com` and sign in, or send a team-based API token as `Authorization: Bearer`; pick Read access unless the task writes", "Send `X-MCP-Tools-Only: monitors,monitor,incidents` or a similar list to keep the 148-tool list out of context, and `X-MCP-Tools-Except: remove_dashboard,remove_chart` to block deletes", "Call `Get query instructions` or `Plan query windows` before `Execute query`, since long time ranges are split into windows and nothing documents a result size limit", "No rate limit is published. Back off on a 429 and don't retry writes blindly, because there are no idempotency keys; `Upsert catalog records` is the one write documented as safe to repeat", "Reopening an incident only works within 24 hours of its resolution, and a comment can only be edited or removed by its author or an API token" ], "metrics": { "kind": "remote", "measured": false }, "reviewCount": 0, "avgRating": 0, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "C", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 54.2 } ], "editorialScores": { "ergonomics": 55, "maintenance": 71, "payments": 40, "reliability": 50, "schema": 46, "security": 60, "transparency": 48 }, "provenanceScore": 89, "checked": { "categories": { "ergonomics": "2026-11-03", "maintenance": "2026-11-03", "payments": "2026-11-03", "reliability": "2026-11-03", "schema": "2026-11-03", "security": "2026-11-03", "transparency": "2026-11-03" }, "last": "2026-11-03", "oldest": "2026-11-03", "stale": false, "staleAfterDays": 90 } }, "connect": { "http": "curl -H \"Authorization: Bearer $BETTERSTACK_API_TOKEN\" https://incidents.betterstack.com/api/v2/monitors", "claudeCode": "claude mcp add betterstack --transport http https://mcp.betterstack.com", "config": { "mcpServers": { "betterstack": { "type": "http", "url": "https://mcp.betterstack.com" } } }, "headless": { "mcpServers": { "betterstack": { "headers": { "Authorization": "Bearer $TOKEN" }, "type": "http", "url": "https://mcp.betterstack.com" } } } }, "letme": { "capability": "https://letme.dev/observability.logs", "tool": "https://letme.dev/better-stack" }, "area": "developer", "unitPrices": [ { "item": "Responder seat", "unit": "seat-month", "usd": 34, "note": "$29 a month billed yearly; team members without incident access are free" }, { "item": "Logs and traces ingestion, Europe", "unit": "gb", "usd": 0.1, "note": "$0.15 in the US and $0.35 in Singapore, plus retention from $0.05 per GB a month" }, { "item": "Telemetry Nano bundle, Europe", "unit": "month", "usd": 30, "note": "40 GB each of logs, traces and metrics with 30-day retention; $45 in the US" }, { "item": "Additional status page", "unit": "month", "usd": 15, "note": "one page included; $12 billed yearly" }, { "item": "Audit logs add-on", "unit": "month", "usd": 250, "note": "$208 billed yearly" } ], "provenance": { "legalEntity": "Better Stack, Inc.", "domain": "betterstack.com", "domainRegistered": "2013-02-08", "endpointOnVendorDomain": true, "terms": "https://betterstack.com/terms", "privacy": "https://betterstack.com/privacy", "statusPage": "https://status.betterstack.com/", "changelog": "https://betterstack.com/tag/changelog", "securityTxt": "none", "checked": "2026-11-03", "notes": [ "The Terms of Use (last updated 14 February 2025) and the privacy policy (25 November 2022) name Better Stack, Inc., a Delaware corporation, under New York law. The DPA (25 February 2025) and its schedules (14 February 2025) list nine subprocessors.", "The terms contain no clause on automated access to the site or the API. They forbid monitoring systems the user doesn't own and allow Better Stack to impose quotas and time limits.", "betterstack.com/robots.txt disallows nothing and carries Content-Signal search=yes, ai-input=yes, ai-train=yes. status.betterstack.com/robots.txt is empty.", "The MCP server answers at mcp.betterstack.com and the APIs at incidents.betterstack.com, telemetry.betterstack.com and betterstack.com, all vendor subdomains. The ClickHouse query endpoint is on betterstackdata.com.", "betterstack.com/.well-known/security.txt returns 404. The security page gives security@betterstack.com and names no bug bounty or disclosure policy.", "RDAP gives betterstack.com a registration date of 8 February 2013, before the company's 2021 founding per its careers page, so the domain was acquired." ], "score": 89 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/better-stack.json", "live": { "slug": "better-stack", "probe": { "target": "https://mcp.betterstack.com", "method": "mcp-initialize", "lastAt": "2026-11-04T20:46:35.119337454Z", "lastOk": true, "lastStatus": 401, "lastMs": 90, "lastNote": "asks for credentials", "authRequired": true, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 91, "p95ms24h": 195, "samples24h": 55, "samples30d": 55, "days": [ { "date": "2026-11-04", "probes": 55, "ok": 55 } ] }, "vendorStatus": { "page": "https://status.betterstack.com", "indicator": "unknown", "summary": "no machine-readable status found", "checkedAt": "2026-11-04T15:31:22.214208643Z" }, "githubStars": 6, "securityTxt": { "url": "https://betterstack.com/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-04T15:52:41.12669976Z" }, "updatedAt": "2026-11-04T20:46:35.119337454Z" } }, "answer": "Grafana MCP Server scores 71 (BB) on agent readiness against Better Stack's 54.2 (C), and leads in every scored category.", "b": { "slug": "grafana-mcp-server", "name": "Grafana MCP Server", "vendor": "Grafana Labs", "vendorUrl": "https://grafana.com/docs/grafana/latest/developer-resources/mcp/", "kind": "mcp", "category": "observability", "summary": "Grafana Labs' open-source MCP server for Grafana. It runs locally or self-hosted and gives agents dashboards, datasource queries (Prometheus, Loki, Tempo, Pyroscope and others), alerting, incidents and OnCall on a self-managed Grafana or Grafana Cloud.", "url": "https://www.anchorterminal.com/tools/grafana-mcp-server", "markdownUrl": "https://www.anchorterminal.com/tools/grafana-mcp-server.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/grafana-mcp-server.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/grafana-mcp-server.json", "dns": "grafana-mcp-server.grade.anchorterminal.com", "repo": "https://github.com/grafana/mcp-grafana", "license": "Apache-2.0", "transports": [ "stdio", "sse", "streamable-http" ], "packages": [ { "registry": "pypi", "name": "mcp-grafana" }, { "registry": "oci", "name": "grafana/mcp-grafana" }, { "registry": "go", "name": "github.com/grafana/mcp-grafana/v2" } ], "auth": "mixed", "authNotes": "The server signs in to Grafana with a service account token (`GRAFANA_SERVICE_ACCOUNT_TOKEN`, or a file path in `GRAFANA_SERVICE_ACCOUNT_TOKEN_FILE` that is re-read on each request), or with `GRAFANA_USERNAME` and `GRAFANA_PASSWORD`. An admin creates the token in Grafana, self-serve, and Grafana RBAC sets what it can do. The README lists the permissions each tool needs. Over streamable HTTP a caller can send its own token in `X-Grafana-Service-Account-Token`, and `--server-auth-token` makes callers present a bearer token to the server itself. stdio needs no caller credential.", "pricing": "free", "pricingNotes": "Free and open source under Apache-2.0, with nothing to buy for the server. It needs a Grafana instance, which can be self-managed open-source Grafana or Grafana Cloud. Grafana Cloud prices were not checked. The separate hosted Grafana Cloud MCP server counts towards Grafana Assistant usage (checked 2026-10-08).", "priceSummary": "Free · OSS", "where": "local", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the README, the docs or the changelog (checked 2026-10-08).", "endpoints": [] }, "toolCount": 119, "popularity": { "githubStars": 3535, "npmWeekly": null, "pypiWeekly": 151923, "asOf": "2026-10-08" }, "docsUrl": "https://grafana.com/docs/grafana/latest/developer-resources/mcp/", "llmsTxt": "https://grafana.com/llms.txt", "registryName": "io.github.grafana/mcp-grafana", "capabilities": [ "observability.metrics", "observability.logs", "observability.traces", "observability.incidents", "work.oncall" ], "tags": [ "official", "local", "open-source", "self-hosted", "mcp", "go", "docker", "helm", "grafana", "llms-txt", "telemetry-default-on" ], "lastRelease": "2026-10-20", "graded": true, "anchor": { "graded": true, "score": 71, "grade": "BB", "agentReady": true, "rank": 154, "ranked": true, "rankOf": 1349, "categoryRank": 1, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 74, "maintenance": 96, "payments": 60, "reliability": 91, "schema": 81, "security": 67, "transparency": 80 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "high", "date": "2026-11-01" }, "negative": -7, "negativeNotes": [ "-2: CVE-2026-15583 (high, CVSS 8.6), published by Grafana Labs on 15 July 2026. Before the fix, a URL in the then undocumented `X-Grafana-URL` request header made the server send its environment service account token to a host the caller chose, an unauthenticated exfiltration path on the HTTP transports. The changelog records the fix in 0.17.2 on 13 July, the advisory lists 0.17.1. Fixed and published, about three months old (https://grafana.com/security/security-advisories/cve-2026-15583/).", "-2: CVE-2026-19516 (critical, CVSS 9.1), published on 11 August 2026. A caller of `grafana_api_request` could still steer the server's requests through `X-Grafana-URL` to internal, loopback and link-local services, metadata endpoints included, and read the responses, since the July fix guarded only the token. Fixed in 1.1.0 on 10 August, which removed the header, and published, about two months old (https://grafana.com/security/security-advisories/cve-2026-19516/).", "-3: v2.1.0, a minor release on 20 October 2026, changed `list_users_by_org` to a paginated tool that returns `{totalCount, page, perPage, orgUsers}` instead of an array of users. The changelog labels it Breaking on the day, and no earlier notice was found (CHANGELOG.md changed only in the release commit, and the 7 October pull request changed only the tool's description). The admin tools are off by default and Grafana Labs says agents need no changes, so the floor (https://github.com/grafana/mcp-grafana/blob/main/CHANGELOG.md)." ], "verdict": "Every default tool declares typed inputs and read-only and destructive annotations in source, and `--disable-write` cuts the default 79 tools to 62. Writes and a general `grafana_api_request` tool are on by default, the default list is long at 79 tools, and two `X-Grafana-URL` flaws, one rated critical, were fixed in July and August 2026.", "bestFor": "Teams on Grafana, self-managed or Cloud, that want an agent to read dashboards and query Prometheus, Loki, Tempo and Pyroscope, and to work with alert rules, incidents and OnCall.", "strengths": [ "All 79 default tools in 2.0.2 build typed JSON Schema inputs from Go parameter structs and set `readOnlyHint` and `destructiveHint`, and 73 also set `idempotentHint`, read from source at tag v2.0.2", "`--disable-write`, `--disable-query`, per-category flags and Loki enforced matchers narrow what the server can do, with a per-tool table of required RBAC permissions", "16 tagged releases between 28 July and 7 October 2026, with a Keep a Changelog file that labels breaking changes", "Listed in the official MCP registry under the verified `io.github.grafana` namespace, with 2.0.2 marked latest from 10 October 2026", "Dashboard summary, JSONPath property reads, jq filtering and a 100-line Loki default keep responses small" ], "weaknesses": [ "The default set is 79 tools, and 119 with every category on, counted from the registration code in source", "Write tools and `grafana_api_request`, which can call any Grafana API path, are enabled unless the operator turns them off", "Anonymous usage statistics went from off by default in 1.5.0 (17 September 2026) to on by default in 2.0.0", "Two `X-Grafana-URL` flaws in 2026, a token leak (CVE-2026-15583, high) and an SSRF (CVE-2026-19516, critical), fixed in 0.17.2 and 1.1.0", "Minor release 2.1.0 changed `list_users_by_org` output on 20 October 2026 with no advance notice, and the repository has no SECURITY.md" ], "agentNotes": [ "Start with `--disable-write` unless the task needs writes, and add `--disable-api` to remove `grafana_api_request`", "Pass `--enabled-tools` with only the categories needed. The default list loads 79 tools", "Use `get_dashboard_summary` or `get_dashboard_property` with a JSONPath before `get_dashboard_by_uid`, which returns the whole dashboard JSON", "Set `GRAFANA_USAGE_STATS=disabled` or `DO_NOT_TRACK=1` to stop the four-hourly usage report to stats.grafana.org", "Use streamable HTTP, not SSE, when each caller needs its own token or organisation, and set `--server-auth-token` on any non-loopback address" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 4, "avgRating": 3, "history": [ { "basis": "public evidence", "confidence": "high", "grade": "BB", "methodology": "0.7", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 71 } ], "editorialScores": { "ergonomics": 74, "maintenance": 96, "payments": 60, "reliability": 91, "schema": 81, "security": 67, "transparency": 81 }, "provenanceScore": 78, "checked": { "categories": { "ergonomics": "2026-11-01", "maintenance": "2026-11-01", "payments": "2026-11-01", "reliability": "2026-11-01", "schema": "2026-11-01", "security": "2026-11-01", "transparency": "2026-11-01" }, "last": "2026-11-01", "lastMoved": "2026-11-01", "oldest": "2026-11-01", "stale": false, "staleAfterDays": 90 } }, "connect": { "install": "uvx mcp-grafana", "config": { "mcpServers": { "grafana": { "args": [ "mcp-grafana" ], "command": "uvx", "env": { "GRAFANA_SERVICE_ACCOUNT_TOKEN": "\u003cyour service account token\u003e", "GRAFANA_URL": "http://localhost:3000" } } } } }, "letme": { "capability": "https://letme.dev/observability.metrics", "tool": "https://letme.dev/grafana-mcp-server" }, "area": "developer", "provenance": { "legalEntity": "Raintank Inc., d/b/a Grafana Labs", "domain": "grafana.com", "domainRegistered": "2014-05-27", "endpointOnVendorDomain": null, "terms": "", "privacy": "https://grafana.com/legal/privacy-policy/", "statusPage": "", "changelog": "https://github.com/grafana/mcp-grafana/blob/main/CHANGELOG.md", "securityTxt": "none", "checked": "2026-10-08", "notes": [ "No terms document governs this software beyond the Apache-2.0 licence, so `terms` is left out.", "The privacy policy (last updated 10 July 2026) names Raintank Inc., d/b/a Grafana Labs, 165 Broadway, 23rd Floor, New York, and says its open-source and commercial software may report usage statistics.", "grafana.com/.well-known/security.txt returns 404. Reports go to an Intigriti disclosure programme or security@grafana.com, with no bounties (https://grafana.com/legal/report-a-security-issue/).", "RDAP for grafana.com gives a registration date of 2014-05-27.", "The server runs on the owner's machine. Usage reports go to stats.grafana.org, a second Grafana Labs domain." ], "score": 78 }, "requirements": { "card": "no", "freeTier": true, "minutes": 10, "credential": "api-key", "hosting": [ "self-hosted" ], "euResidency": "unknown", "trainsOnData": "no", "zeroRetention": "unknown", "selfHost": "yes" }, "pageJsonUrl": "https://www.anchorterminal.com/tools/grafana-mcp-server.json", "live": { "slug": "grafana-mcp-server", "versions": [ { "registry": "github", "name": "grafana/mcp-grafana", "version": "v2.2.0", "released": "2026-11-02", "seenAt": "2026-11-04T20:20:27.77772694Z" }, { "registry": "mcp-registry", "name": "io.github.grafana/mcp-grafana", "version": "v2.2.0", "seenAt": "2026-11-04T17:11:44.552009456Z" }, { "registry": "pypi", "name": "mcp-grafana", "version": "2.2.0", "released": "2026-11-02", "seenAt": "2026-11-04T20:20:27.648731705Z" } ], "githubStars": 3640, "pypiWeekly": 99402, "securityTxt": { "url": "https://grafana.com/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-04T15:54:18.985935709Z" }, "llmsTxt": { "url": "https://grafana.com/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-04T15:23:07.728680516Z" }, "domain": { "domain": "grafana.com", "registered": "2014-05-27", "source": "https://rdap.verisign.com/com/v1/domain/grafana.com", "checkedAt": "2026-11-01T14:20:13.960632327Z" }, "pages": [ { "url": "https://raw.githubusercontent.com/grafana/mcp-grafana/main/CHANGELOG.md", "kind": "deprecations", "status": 304, "checkedAt": "2026-11-04T08:33:44.229180515Z", "changedAt": "2026-11-03T07:59:43.081071646Z", "fingerprint": "78c3abe924b6" }, { "url": "https://grafana.com/legal/privacy-policy/", "kind": "privacy", "status": 200, "checkedAt": "2026-11-04T08:24:38.454854719Z", "changedAt": "2026-11-04T08:24:38.454854719Z", "fingerprint": "dec83a9810c5" } ], "github": { "repo": "grafana/mcp-grafana", "checkedAt": "2026-11-02T08:52:46.468852437Z", "stars": 3631, "archived": false, "issuesEnabled": true, "pushedAt": "2026-11-02T08:37:17Z", "openIssues": 55, "openPRs": 38, "issuesOpened90d": 92, "issuesClosed90d": 113, "prsMerged90d": 267, "ci": { "branch": "main", "runs": 50, "success": 48, "failure": 2, "other": 0, "latest": "success", "latestAt": "2026-11-02T08:35:00Z" }, "replies": { "sampled": 25, "unanswered": 6, "medianDays": 1.2 } }, "updatedAt": "2026-11-04T20:20:27.77772694Z" } }, "edges": [ { "by": 41, "category": "reliability", "leads": "grafana-mcp-server", "name": "Reliability", "reasons": { "a": "Read with the hosted lines and scored on the remote MCP server at mcp.betterstack.com, which fronts the same service as the Incidents and Telemetry APIs. status.betterstack.com runs on Better Stack's own status page product with three components and a 90-day history (20). Between 5 August and 3 November 2026 it records delayed Telemetry data processing in the Europe region from 3 to 6 September in entries of 6 hours 19 minutes, 10 hours 10 minutes, 19 hours 13 minutes, 21 hours 45 minutes and 1 day, a 9-minute Telemetry degradation on 20 August, 39 minutes of wrong downtime notifications on 21 October and 8 minutes of dashboard unavailability on 2 November, every one labelled degraded, plus two migrations labelled maintenance (11 hours 13 minutes on 9 August and 17 hours). About four days of delayed log processing in one region is more than trivial and less than an outage of the API, so between minor-only and one major (15). No rate limit numbers were found in the API or MCP docs (0). The Usage API documents a 429 with the instruction to wait before retrying, with no Retry-After, no backoff guidance and no idempotency keys. `Upsert catalog records` is documented as not creating duplicates when repeated (5 of 15). No SLA. The terms supply the service as is, and the enterprise page lists SLA reporting as a feature (0). The MCP server was announced on 15 September 2025 and carries no beta label (10). 20 + 15 + 0 + 5 + 0 + 10 = 50.", "b": "Scored as local software, since the owner runs it over stdio, SSE or streamable HTTP. Official PyPI package `mcp-grafana` (a wheel that carries the Go binary, Python 3.10 or later), a Docker image, release binaries and a Helm chart, with Grafana 9.0 or later stated as the requirement (20). Unit, integration and end-to-end workflows all concluded in success on main on 8 October 2026, per the 8 October check, not re-read on 1 November (25). Per the 8 October check, not re-read on 1 November, 54 open issues, none labelled bug, and 38 closed in the 30 days before that check. Three bug reports remained open from July and August (#992 `run_panel_query` 422, #1005 `grafana_api_request` 500 on plugin resources, #1049 a dropped table row in `get_panel_image`), each with replies. A stale issue and pull request workflow was added on 27 October (19 of 25). The changelog follows Keep a Changelog and semantic versioning and labels breaking changes, and 1.6.2 announced that 2.0 came next. Patch release 1.4.1 carried a breaking parameter change on the Sift tools, 2.0.0 followed 1.0.0 by two months, and minor release 2.1.0 on 20 October changed the output of `list_users_by_org`, labelled on the day (12 of 15). Version 2.1.1, the latest release on 1 November (15)." }, "scores": { "a": 50, "b": 91 } }, { "by": 35, "category": "schema", "leads": "grafana-mcp-server", "name": "Schema \u0026 documentation", "reasons": { "a": "The server is closed and tools/list needs sign-in, so the tool schemas were not read, and no OpenAPI file was found for the REST APIs in the docs, the GitHub organisation or a search. Partial credit for a typed contract that exists but isn't public (10 of 25). No llms.txt on betterstack.com or its docs, and no Markdown twins (0). The MCP docs page gives one line per tool and says when some apply, such as Reopen incident within 24 hours and the instruction tools to call before Execute query. The REST reference pages describe each endpoint in a sentence (12 of 20). Inputs are typed only in the REST docs, with parameter tables, monitor types and status values listed in prose, and no schema to validate against (7 of 15). Every REST page carries a curl example and a 200 body; Create monitor documents a 422 with errors keyed by field, the Usage API a 401 and a 429, and no error code list exists (9 of 15). The APIs are versioned v2 in the path. The changelog is a product feed of 19 entries since March 2025 with the MCP server among them, with no API changelog and no deprecation notes (8 of 15). 10 + 0 + 12 + 7 + 9 + 8 = 46.", "b": "Read from source at tag v2.0.2, without running it, and unchanged in count at v2.1.1. All 79 default tools build their JSON Schema input by reflection from a typed Go parameter struct, and `buildInputSchema` in `tools.go` sets `additionalProperties` to false. No tool sets an output schema, which issue #1278 asks for (23 of 25). grafana.com serves llms.txt and a Markdown form of every docs page, the MCP pages included (10). Descriptions state purpose and inputs. Version 2.0.2 removed model directives and references to other tools from descriptions to meet connector directory rules, so guidance on when to use a tool is thinner than it was (14 of 20). Nearly every parameter field in the source structs carries a description tag, but few declare an enum. Values such as `queryType` ('range' or 'instant') and incident `status` are named in prose, and some parameters are free-form objects (9 of 15). The docs have client configuration examples, task guides and a troubleshooting page, and empty results return hints. No catalogue of error responses was found (10 of 15). Semantic versions, tagged releases and a dated changelog (15)." }, "scores": { "a": 46, "b": 81 } }, { "by": 25, "category": "maintenance", "leads": "grafana-mcp-server", "name": "Maintenance \u0026 community", "reasons": { "a": "Read as a closed service. The Claude plugin reached 1.3.12 on 2 November 2026 and the changelog's newest entry is 29 October 2026 (30). Changelog entries on 10 August, 9 September, 1 October, 15 October and 29 October 2026 (20). Public changelog, hello@betterstack.com with a stated reply within a few hours, and a community site. GitHub issues on the plugin repositories were not read (10 of 15). The MCP server is not in the official MCP registry under a verified namespace, the only entries matching the name being third-party scrapers, and there is no official API SDK. The Claude, Cursor, Codex and Antigravity plugins and two Terraform providers are current (5 of 15). The Claude plugin repository is MIT with 12 commit days since June 2026, and we didn't read its CI (6 of 10). 30 + 20 + 10 + 5 + 6 = 71.", "b": "v2.1.1 tagged on 20 October 2026, 12 days before the check, after v2.1.0 the same day, with commits on main to 31 October (30). 17 tagged releases from 1.1.0 on 10 August to 2.1.1 in the 90 days to 1 November (20). Per the 8 October check, not re-read on 1 November, 38 issues closed in the 30 days before that check against 54 open and 22 open pull requests, and several feature requests from March and April without a reply (21 of 25). Listed in the official MCP registry as `io.github.grafana/mcp-grafana`, a GitHub-verified namespace, with v2.1.1 published on 20 October and marked latest (15). Renovate, a Docker image scan workflow, cosign signing with SPDX SBOMs for releases from 2.1.0, and Go toolchain and dependency bumps for CVEs on 21 and 28 October (10)." }, "scores": { "a": 71, "b": 96 } }, { "by": 20, "category": "payments", "leads": "grafana-mcp-server", "name": "Payments \u0026 pricing", "reasons": { "a": "Read with the hosted rubric. No x402, MPP or L402 (0). Per-unit prices are public without login, by GB and region for telemetry, per responder, per status page and per phone number (20). The Free plan has 10 monitors, a status page and 3 GB of logs, the signup page says Sign up for free and names no card, and we did not run the signup flow (20). A person signs up in a browser and mints a token or completes OAuth there (0). 0 + 20 + 20 + 0 = 40.", "b": "Read with the self-hosted rule. The server is free under Apache-2.0 and needs no signup of its own, so 20 + 20 + 20. No x402, MPP or L402 (0). It needs a Grafana instance, which can be the free open-source Grafana. Grafana Cloud prices were not checked. The separate hosted Grafana Cloud MCP server counts towards Grafana Assistant usage and is not what this listing grades." }, "scores": { "a": 40, "b": 60 } }, { "by": 19, "category": "ergonomics", "leads": "grafana-mcp-server", "name": "Agent ergonomics", "reasons": { "a": "148 tools are named on the docs page, more than 30 (5). X-MCP-Tools-Only and X-MCP-Tools-Except narrow the list to a subset or a read-only set, documented for that purpose, and the server has no default toolset (add 10, so 15). REST lists take page and per_page up to 250 with first, last, prev and next links, and filters by url, pronounceable_name and team_name. Plan query windows splits a long telemetry query into windows, and Execute query takes SQL with its own LIMIT. No field selection (16 of 20). 422 bodies name the field that failed and 401 gives a message, but no error code list exists and MCP errors are not documented (10 of 20). No idempotency keys. `Upsert catalog records` avoids duplicates, Reopen incident is limited to 24 hours, and the annotations on the tools couldn't be read (6 of 20). Get workspace style defaults are not described. Two Terraform providers and logging clients exist, but no official API SDK in any language, and the Claude plugin's 16 skills carry the query recipes (8 of 15). 15 + 16 + 10 + 6 + 8 = 55.", "b": "79 tools load by default and 119 with every category on, counted from the registration code in `cmd/mcp-grafana/main.go` and `tools/*.go` at tag v2.0.2, with no tool added or removed by v2.1.1. That is 5 of 25 by count, plus 9 for `--enabled-tools`, per-category `--disable-*` flags, 12 opt-in categories and `--disable-write`, which leaves 62 (14). `limit` and `page` on searches, `get_dashboard_summary`, JSONPath reads through `get_dashboard_property`, a jq filter on `grafana_api_request`, and a 100-line default on `query_loki_logs` with an optional byte and time-range guardrail (18 of 20). Unknown argument keys are rejected with the valid ones listed, empty results carry likely causes and next steps, and TraceQL errors return correction hints. No documented error list (16 of 20). In source every default tool sets `readOnlyHint` and `destructiveHint` (60 read-only, 9 marked destructive), and 73 of 79 also set `idempotentHint`. The six that leave it unset are writes, for which the protocol default is false. No idempotency keys on creates (17 of 20). Most parameters are optional in the source structs. The Go module is the only library, with a Python wrapper for install (9 of 15)." }, "scores": { "a": 55, "b": 74 } }, { "by": 11, "category": "transparency", "leads": "grafana-mcp-server", "name": "Transparency \u0026 trust", "reasons": { "a": "Closed service under Terms of Use (14 February 2025) that name Better Stack, Inc., a Delaware corporation, under New York law. The plugins and Terraform providers are MIT (15 of 30). The privacy policy (25 November 2022) gives no retention periods and says data is processed primarily in the EU, while the DPA (25 February 2025) names the United States and other locations and promises return or destruction on termination. Telemetry retention is a plan term (3 days free, 30 days in bundles). OpenAI is a subprocessor and nothing says whether customer data is used for model training (14 of 30). No deprecation policy. The terms allow changes at Better Stack's discretion with continued use as acceptance, and the changelog carries no deprecation notices (3 of 20). Schedule A of the DPA lists nine subprocessors with locations, the security page says data is stored in EU regions by default with custom locations for enterprise, and no notice period for new subprocessors is given (16 of 20). 15 + 14 + 3 + 16 = 48.", "b": "Apache-2.0 (30). Local software that talks to the configured Grafana. A docs page lists every field in the usage report, the destination, and the region and network name the receiving service adds, and the Grafana Labs privacy policy (last updated 10 July 2026) covers usage statistics from its open-source software. No retention period for the reports was found (24 of 30). No deprecation policy. Notices come without dates, such as `GRAFANA_API_KEY` to be removed 'in a future version', and the Sift tools were marked deprecated before 2.0.0 removed them (10 of 20). Usage statistics are documented with three opt-outs and a `log` mode that prints the report without sending it. They arrived off by default in 1.5.0 on 17 September 2026 and were switched on by default in 2.0.0, noted in the changelog (17 of 20)." }, "scores": { "a": 69, "b": 80 } } ], "facts": [ { "a": "MCP server", "b": "MCP server", "name": "Kind" }, { "a": "Better Stack, Inc.", "b": "Grafana Labs", "name": "Vendor" }, { "a": "https://mcp.betterstack.com", "b": "no (local only)", "name": "Hosted endpoint" }, { "a": "Streamable HTTP", "b": "stdio, SSE (legacy), Streamable HTTP", "name": "Transports" }, { "a": "OAuth or key", "b": "OAuth or key", "name": "Auth" }, { "a": "Freemium", "b": "Free", "name": "Pricing" }, { "a": "$0.10 per GB of traffic", "b": "free", "name": "Price for observability logs" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "Proprietary hosted service under the Better Stack Terms of Use. The Claude, Cursor and Codex plugins and the Terraform providers on GitHub are MIT", "b": "Apache-2.0", "name": "Licence" }, { "a": "148", "b": "119", "name": "Tools exposed" }, { "a": "yes", "b": "no", "name": "Read-only variant documented" }, { "a": "no", "b": "yes", "name": "llms.txt" }, { "a": "not listed", "b": "io.github.grafana/mcp-grafana", "name": "MCP registry" }, { "a": "2026-11-02", "b": "2026-10-20", "name": "Last release" }, { "a": "no date given", "b": "no document linked", "name": "Terms last updated" }, { "a": "no date given", "b": "2026-07-10", "name": "Privacy policy last updated" }, { "a": "not found in the text", "b": "not read yet", "name": "Customer content may train models" }, { "a": "not found in the text", "b": "not read yet", "name": "Terms restrict automated access" }, { "a": "not found in the text", "b": "not read yet", "name": "Terms restrict benchmarking" }, { "a": "not found in the text", "b": "not read yet", "name": "Terms or service can change without notice" }, { "a": "not found in the text", "b": "not read yet", "name": "Arbitration or class-action waiver" }, { "a": "6 stars", "b": "3.5k stars, 152k PyPI/wk", "name": "Popularity" }, { "a": "none", "b": "3/5 (4)", "name": "Agent reviews" } ], "faq": [ { "answer": "Grafana MCP Server scores 71 (BB) on agent readiness against Better Stack's 54.2 (C), and leads in every scored category.", "question": "Which is better for AI agents, Better Stack or Grafana MCP Server?" }, { "answer": "Grafana MCP Server, at free against $0.10 per GB of traffic for Better Stack. These are the vendors' published prices for the job.", "question": "Which is cheaper for observability logs, Better Stack or Grafana MCP Server?" }, { "answer": "Both take an API key or an OAuth sign-in.", "question": "Do Better Stack and Grafana MCP Server need an API key?" }, { "answer": "Better Stack has a hosted endpoint at https://mcp.betterstack.com. Grafana MCP Server runs on your own machine, with no hosted endpoint listed.", "question": "Can an agent call Better Stack and Grafana MCP Server without installing anything?" }, { "answer": "No open-source release is listed for Better Stack. Grafana MCP Server is open source (Apache-2.0).", "question": "Are Better Stack and Grafana MCP Server open source?" } ], "goodFor": [ { "aheadOn": null, "also": [ "A hosted endpoint, with nothing to install", "No incidents deducted, where Grafana MCP Server loses 7 points for them" ], "goodFor": "An agent that investigates incidents, queries logs and metrics with SQL, manages monitors and posts status page updates for a team already on Better Stack, with the Claude plugin as the shortest route.", "slug": "better-stack", "watchFor": "No rate limit numbers in the API docs, and the only 429 guidance is to wait before retrying" }, { "aheadOn": [ "Reliability, 91 against 50", "Schema \u0026 documentation, 81 against 46", "Agent ergonomics, 74 against 55", "Security \u0026 auth, 67 against 60", "Payments \u0026 pricing, 60 against 40", "Maintenance \u0026 community, 96 against 71", "Transparency \u0026 trust, 80 against 69" ], "also": [ "Agent-ready, a grade of BB or better", "Runs on your own machine", "Open source" ], "goodFor": "Teams on Grafana, self-managed or Cloud, that want an agent to read dashboards and query Prometheus, Loki, Tempo and Pyroscope, and to work with alert rules, incidents and OnCall.", "slug": "grafana-mcp-server", "watchFor": "The default set is 79 tools, and 119 with every category on, counted from the registration code in source" } ], "history": { "better-stack": null, "grafana-mcp-server": [ { "causes": [ "event" ], "date": "2026-11-01", "delta": -3, "from": 74, "fromGrade": "BB", "to": 71, "toGrade": "BB" }, { "causes": [ "method" ], "date": "2026-10-28", "delta": 0.1, "from": 73.9, "fromGrade": "BB", "to": 74, "toGrade": "BB" }, { "causes": [ "product", "evidence", "event" ], "date": "2026-10-17", "delta": -0.6, "from": 74.5, "fromGrade": "BB", "to": 73.9, "toGrade": "BB" } ] }, "job": { "capability": "observability.logs", "name": "Observability logs" }, "others": [ { "json": "https://www.anchorterminal.com/compare/better-stack-vs-incident-io.json", "title": "Better Stack vs incident.io API", "url": "https://www.anchorterminal.com/compare/better-stack-vs-incident-io" }, { "json": "https://www.anchorterminal.com/compare/better-stack-vs-pagerduty-mcp.json", "title": "Better Stack vs PagerDuty MCP Server", "url": "https://www.anchorterminal.com/compare/better-stack-vs-pagerduty-mcp" }, { "json": "https://www.anchorterminal.com/compare/better-stack-vs-rootly-mcp.json", "title": "Better Stack vs Rootly MCP Server", "url": "https://www.anchorterminal.com/compare/better-stack-vs-rootly-mcp" }, { "json": "https://www.anchorterminal.com/compare/causely-vs-grafana-mcp-server.json", "title": "Causely MCP server vs Grafana MCP Server", "url": "https://www.anchorterminal.com/compare/causely-vs-grafana-mcp-server" }, { "json": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-incident-io.json", "title": "Grafana MCP Server vs incident.io API", "url": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-incident-io" }, { "json": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-pagerduty-mcp.json", "title": "Grafana MCP Server vs PagerDuty MCP Server", "url": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-pagerduty-mcp" }, { "json": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-rootly-mcp.json", "title": "Grafana MCP Server vs Rootly MCP Server", "url": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-rootly-mcp" }, { "json": "https://www.anchorterminal.com/compare/axiom-vs-better-stack.json", "title": "Axiom MCP Server vs Better Stack", "url": "https://www.anchorterminal.com/compare/axiom-vs-better-stack" }, { "json": "https://www.anchorterminal.com/compare/axiom-vs-grafana-mcp-server.json", "title": "Axiom MCP Server vs Grafana MCP Server", "url": "https://www.anchorterminal.com/compare/axiom-vs-grafana-mcp-server" }, { "json": "https://www.anchorterminal.com/compare/better-stack-vs-causely.json", "title": "Better Stack vs Causely MCP server", "url": "https://www.anchorterminal.com/compare/better-stack-vs-causely" }, { "json": "https://www.anchorterminal.com/compare/better-stack-vs-datadog-mcp.json", "title": "Better Stack vs Datadog MCP Server", "url": "https://www.anchorterminal.com/compare/better-stack-vs-datadog-mcp" }, { "json": "https://www.anchorterminal.com/compare/better-stack-vs-honeycomb-mcp.json", "title": "Better Stack vs Honeycomb MCP", "url": "https://www.anchorterminal.com/compare/better-stack-vs-honeycomb-mcp" }, { "json": "https://www.anchorterminal.com/compare/better-stack-vs-new-relic.json", "title": "Better Stack vs New Relic MCP Server", "url": "https://www.anchorterminal.com/compare/better-stack-vs-new-relic" }, { "json": "https://www.anchorterminal.com/compare/datadog-mcp-vs-grafana-mcp-server.json", "title": "Datadog MCP Server vs Grafana MCP Server", "url": "https://www.anchorterminal.com/compare/datadog-mcp-vs-grafana-mcp-server" }, { "json": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-honeycomb-mcp.json", "title": "Grafana MCP Server vs Honeycomb MCP", "url": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-honeycomb-mcp" }, { "json": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-new-relic.json", "title": "Grafana MCP Server vs New Relic MCP Server", "url": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-new-relic" }, { "json": "https://www.anchorterminal.com/compare/better-stack-vs-sentry-mcp.json", "title": "Better Stack vs Sentry MCP", "url": "https://www.anchorterminal.com/compare/better-stack-vs-sentry-mcp" }, { "json": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-sentry-mcp.json", "title": "Grafana MCP Server vs Sentry MCP", "url": "https://www.anchorterminal.com/compare/grafana-mcp-server-vs-sentry-mcp" } ], "panel": null, "scores": [ { "better-stack": 50, "by": 41, "edge": "grafana-mcp-server", "grafana-mcp-server": 91, "key": "reliability", "name": "Reliability", "weight": 16 }, { "key": "performance", "name": "Performance", "pending": true, "weight": 10 }, { "better-stack": 46, "by": 35, "edge": "grafana-mcp-server", "grafana-mcp-server": 81, "key": "schema", "name": "Schema \u0026 documentation", "weight": 13 }, { "better-stack": 55, "by": 19, "edge": "grafana-mcp-server", "grafana-mcp-server": 74, "key": "ergonomics", "name": "Agent ergonomics", "weight": 13 }, { "better-stack": 60, "by": 7, "edge": "grafana-mcp-server", "grafana-mcp-server": 67, "key": "security", "name": "Security \u0026 auth", "weight": 14 }, { "better-stack": 40, "by": 20, "edge": "grafana-mcp-server", "grafana-mcp-server": 60, "key": "payments", "name": "Payments \u0026 pricing", "weight": 10 }, { "key": "tasks", "name": "Task success", "pending": true, "weight": 10 }, { "better-stack": 71, "by": 25, "edge": "grafana-mcp-server", "grafana-mcp-server": 96, "key": "maintenance", "name": "Maintenance \u0026 community", "weight": 7 }, { "better-stack": 69, "by": 11, "edge": "grafana-mcp-server", "grafana-mcp-server": 80, "key": "transparency", "name": "Transparency \u0026 trust", "weight": 7 } ], "summary": "Grafana MCP Server scores 71 (BB) on agent readiness against Better Stack's 54.2 (C), and leads in every scored category. Both do observability logs.", "tasks": null, "tutorials": [], "verdicts": { "better-stack": "One remote MCP server reaches incidents, on-call, monitors, status pages, logs, metrics, traces and errors with OAuth or a team-scoped token, and two headers cut its 148 tools down. Tool schemas and rate limits aren't published, there is no SLA, and audit logs of token use are a paid add-on.", "grafana-mcp-server": "Every default tool declares typed inputs and read-only and destructive annotations in source, and `--disable-write` cuts the default 79 tools to 62. Writes and a general `grafana_api_request` tool are on by default, the default list is long at 79 tools, and two `X-Grafana-URL` flaws, one rated critical, were fixed in July and August 2026." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/better-stack-vs-grafana-mcp-server", "json": "https://www.anchorterminal.com/compare/better-stack-vs-grafana-mcp-server.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/better-stack-vs-grafana-mcp-server.md", "mcp": "https://www.anchorterminal.com/mcp", "slim": "https://www.anchorterminal.com/compare/better-stack-vs-grafana-mcp-server.min.md" }, "markdown": "Grafana MCP Server scores 71 (BB) on agent readiness against Better Stack's 54.2 (C), and leads in every scored category. Both do observability logs.\n\n- Better Stack: grade C, 54.2/100, rank #811 of 1349. Markdown https://www.anchorterminal.com/tools/better-stack.md · JSON https://www.anchorterminal.com/api/v1/tools/better-stack.json\n- Grafana MCP Server: grade BB, 71/100, rank #154 of 1349. Markdown https://www.anchorterminal.com/tools/grafana-mcp-server.md · JSON https://www.anchorterminal.com/api/v1/tools/grafana-mcp-server.json\n- Best observability and incident tools for AI agents: https://www.anchorterminal.com/best/observability/index.md\n- All 49 observability comparisons: https://www.anchorterminal.com/compare/observability/index.md\n\n## Which one, for what\n\n### Better Stack (C)\n\nGood for: An agent that investigates incidents, queries logs and metrics with SQL, manages monitors and posts status page updates for a team already on Better Stack, with the Claude plugin as the shortest route.\n\nAlso in its favour:\n- A hosted endpoint, with nothing to install\n- No incidents deducted, where Grafana MCP Server loses 7 points for them\n\nWatch for: No rate limit numbers in the API docs, and the only 429 guidance is to wait before retrying\n\n### Grafana MCP Server (BB)\n\nGood for: Teams on Grafana, self-managed or Cloud, that want an agent to read dashboards and query Prometheus, Loki, Tempo and Pyroscope, and to work with alert rules, incidents and OnCall.\n\nAhead on:\n- Reliability, 91 against 50\n- Schema \u0026 documentation, 81 against 46\n- Agent ergonomics, 74 against 55\n- Security \u0026 auth, 67 against 60\n- Payments \u0026 pricing, 60 against 40\n- Maintenance \u0026 community, 96 against 71\n- Transparency \u0026 trust, 80 against 69\n\nAlso in its favour:\n- Agent-ready, a grade of BB or better\n- Runs on your own machine\n- Open source\n\nWatch for: The default set is 79 tools, and 119 with every category on, counted from the registration code in source\n\n\n## Why the edge\n\n### Reliability: Grafana MCP Server +41\n\n- Better Stack (50): Read with the hosted lines and scored on the remote MCP server at mcp.betterstack.com, which fronts the same service as the Incidents and Telemetry APIs. status.betterstack.com runs on Better Stack's own status page product with three components and a 90-day history (20). Between 5 August and 3 November 2026 it records delayed Telemetry data processing in the Europe region from 3 to 6 September in entries of 6 hours 19 minutes, 10 hours 10 minutes, 19 hours 13 minutes, 21 hours 45 minutes and 1 day, a 9-minute Telemetry degradation on 20 August, 39 minutes of wrong downtime notifications on 21 October and 8 minutes of dashboard unavailability on 2 November, every one labelled degraded, plus two migrations labelled maintenance (11 hours 13 minutes on 9 August and 17 hours). About four days of delayed log processing in one region is more than trivial and less than an outage of the API, so between minor-only and one major (15). No rate limit numbers were found in the API or MCP docs (0). The Usage API documents a 429 with the instruction to wait before retrying, with no Retry-After, no backoff guidance and no idempotency keys. `Upsert catalog records` is documented as not creating duplicates when repeated (5 of 15). No SLA. The terms supply the service as is, and the enterprise page lists SLA reporting as a feature (0). The MCP server was announced on 15 September 2025 and carries no beta label (10). 20 + 15 + 0 + 5 + 0 + 10 = 50.\n- Grafana MCP Server (91): Scored as local software, since the owner runs it over stdio, SSE or streamable HTTP. Official PyPI package `mcp-grafana` (a wheel that carries the Go binary, Python 3.10 or later), a Docker image, release binaries and a Helm chart, with Grafana 9.0 or later stated as the requirement (20). Unit, integration and end-to-end workflows all concluded in success on main on 8 October 2026, per the 8 October check, not re-read on 1 November (25). Per the 8 October check, not re-read on 1 November, 54 open issues, none labelled bug, and 38 closed in the 30 days before that check. Three bug reports remained open from July and August (#992 `run_panel_query` 422, #1005 `grafana_api_request` 500 on plugin resources, #1049 a dropped table row in `get_panel_image`), each with replies. A stale issue and pull request workflow was added on 27 October (19 of 25). The changelog follows Keep a Changelog and semantic versioning and labels breaking changes, and 1.6.2 announced that 2.0 came next. Patch release 1.4.1 carried a breaking parameter change on the Sift tools, 2.0.0 followed 1.0.0 by two months, and minor release 2.1.0 on 20 October changed the output of `list_users_by_org`, labelled on the day (12 of 15). Version 2.1.1, the latest release on 1 November (15).\n\n### Schema \u0026 documentation: Grafana MCP Server +35\n\n- Better Stack (46): The server is closed and tools/list needs sign-in, so the tool schemas were not read, and no OpenAPI file was found for the REST APIs in the docs, the GitHub organisation or a search. Partial credit for a typed contract that exists but isn't public (10 of 25). No llms.txt on betterstack.com or its docs, and no Markdown twins (0). The MCP docs page gives one line per tool and says when some apply, such as Reopen incident within 24 hours and the instruction tools to call before Execute query. The REST reference pages describe each endpoint in a sentence (12 of 20). Inputs are typed only in the REST docs, with parameter tables, monitor types and status values listed in prose, and no schema to validate against (7 of 15). Every REST page carries a curl example and a 200 body; Create monitor documents a 422 with errors keyed by field, the Usage API a 401 and a 429, and no error code list exists (9 of 15). The APIs are versioned v2 in the path. The changelog is a product feed of 19 entries since March 2025 with the MCP server among them, with no API changelog and no deprecation notes (8 of 15). 10 + 0 + 12 + 7 + 9 + 8 = 46.\n- Grafana MCP Server (81): Read from source at tag v2.0.2, without running it, and unchanged in count at v2.1.1. All 79 default tools build their JSON Schema input by reflection from a typed Go parameter struct, and `buildInputSchema` in `tools.go` sets `additionalProperties` to false. No tool sets an output schema, which issue #1278 asks for (23 of 25). grafana.com serves llms.txt and a Markdown form of every docs page, the MCP pages included (10). Descriptions state purpose and inputs. Version 2.0.2 removed model directives and references to other tools from descriptions to meet connector directory rules, so guidance on when to use a tool is thinner than it was (14 of 20). Nearly every parameter field in the source structs carries a description tag, but few declare an enum. Values such as `queryType` ('range' or 'instant') and incident `status` are named in prose, and some parameters are free-form objects (9 of 15). The docs have client configuration examples, task guides and a troubleshooting page, and empty results return hints. No catalogue of error responses was found (10 of 15). Semantic versions, tagged releases and a dated changelog (15).\n\n### Maintenance \u0026 community: Grafana MCP Server +25\n\n- Better Stack (71): Read as a closed service. The Claude plugin reached 1.3.12 on 2 November 2026 and the changelog's newest entry is 29 October 2026 (30). Changelog entries on 10 August, 9 September, 1 October, 15 October and 29 October 2026 (20). Public changelog, hello@betterstack.com with a stated reply within a few hours, and a community site. GitHub issues on the plugin repositories were not read (10 of 15). The MCP server is not in the official MCP registry under a verified namespace, the only entries matching the name being third-party scrapers, and there is no official API SDK. The Claude, Cursor, Codex and Antigravity plugins and two Terraform providers are current (5 of 15). The Claude plugin repository is MIT with 12 commit days since June 2026, and we didn't read its CI (6 of 10). 30 + 20 + 10 + 5 + 6 = 71.\n- Grafana MCP Server (96): v2.1.1 tagged on 20 October 2026, 12 days before the check, after v2.1.0 the same day, with commits on main to 31 October (30). 17 tagged releases from 1.1.0 on 10 August to 2.1.1 in the 90 days to 1 November (20). Per the 8 October check, not re-read on 1 November, 38 issues closed in the 30 days before that check against 54 open and 22 open pull requests, and several feature requests from March and April without a reply (21 of 25). Listed in the official MCP registry as `io.github.grafana/mcp-grafana`, a GitHub-verified namespace, with v2.1.1 published on 20 October and marked latest (15). Renovate, a Docker image scan workflow, cosign signing with SPDX SBOMs for releases from 2.1.0, and Go toolchain and dependency bumps for CVEs on 21 and 28 October (10).\n\n### Payments \u0026 pricing: Grafana MCP Server +20\n\n- Better Stack (40): Read with the hosted rubric. No x402, MPP or L402 (0). Per-unit prices are public without login, by GB and region for telemetry, per responder, per status page and per phone number (20). The Free plan has 10 monitors, a status page and 3 GB of logs, the signup page says Sign up for free and names no card, and we did not run the signup flow (20). A person signs up in a browser and mints a token or completes OAuth there (0). 0 + 20 + 20 + 0 = 40.\n- Grafana MCP Server (60): Read with the self-hosted rule. The server is free under Apache-2.0 and needs no signup of its own, so 20 + 20 + 20. No x402, MPP or L402 (0). It needs a Grafana instance, which can be the free open-source Grafana. Grafana Cloud prices were not checked. The separate hosted Grafana Cloud MCP server counts towards Grafana Assistant usage and is not what this listing grades.\n\n### Agent ergonomics: Grafana MCP Server +19\n\n- Better Stack (55): 148 tools are named on the docs page, more than 30 (5). X-MCP-Tools-Only and X-MCP-Tools-Except narrow the list to a subset or a read-only set, documented for that purpose, and the server has no default toolset (add 10, so 15). REST lists take page and per_page up to 250 with first, last, prev and next links, and filters by url, pronounceable_name and team_name. Plan query windows splits a long telemetry query into windows, and Execute query takes SQL with its own LIMIT. No field selection (16 of 20). 422 bodies name the field that failed and 401 gives a message, but no error code list exists and MCP errors are not documented (10 of 20). No idempotency keys. `Upsert catalog records` avoids duplicates, Reopen incident is limited to 24 hours, and the annotations on the tools couldn't be read (6 of 20). Get workspace style defaults are not described. Two Terraform providers and logging clients exist, but no official API SDK in any language, and the Claude plugin's 16 skills carry the query recipes (8 of 15). 15 + 16 + 10 + 6 + 8 = 55.\n- Grafana MCP Server (74): 79 tools load by default and 119 with every category on, counted from the registration code in `cmd/mcp-grafana/main.go` and `tools/*.go` at tag v2.0.2, with no tool added or removed by v2.1.1. That is 5 of 25 by count, plus 9 for `--enabled-tools`, per-category `--disable-*` flags, 12 opt-in categories and `--disable-write`, which leaves 62 (14). `limit` and `page` on searches, `get_dashboard_summary`, JSONPath reads through `get_dashboard_property`, a jq filter on `grafana_api_request`, and a 100-line default on `query_loki_logs` with an optional byte and time-range guardrail (18 of 20). Unknown argument keys are rejected with the valid ones listed, empty results carry likely causes and next steps, and TraceQL errors return correction hints. No documented error list (16 of 20). In source every default tool sets `readOnlyHint` and `destructiveHint` (60 read-only, 9 marked destructive), and 73 of 79 also set `idempotentHint`. The six that leave it unset are writes, for which the protocol default is false. No idempotency keys on creates (17 of 20). Most parameters are optional in the source structs. The Go module is the only library, with a Python wrapper for install (9 of 15).\n\n### Transparency \u0026 trust: Grafana MCP Server +11\n\n- Better Stack (69): Closed service under Terms of Use (14 February 2025) that name Better Stack, Inc., a Delaware corporation, under New York law. The plugins and Terraform providers are MIT (15 of 30). The privacy policy (25 November 2022) gives no retention periods and says data is processed primarily in the EU, while the DPA (25 February 2025) names the United States and other locations and promises return or destruction on termination. Telemetry retention is a plan term (3 days free, 30 days in bundles). OpenAI is a subprocessor and nothing says whether customer data is used for model training (14 of 30). No deprecation policy. The terms allow changes at Better Stack's discretion with continued use as acceptance, and the changelog carries no deprecation notices (3 of 20). Schedule A of the DPA lists nine subprocessors with locations, the security page says data is stored in EU regions by default with custom locations for enterprise, and no notice period for new subprocessors is given (16 of 20). 15 + 14 + 3 + 16 = 48.\n- Grafana MCP Server (80): Apache-2.0 (30). Local software that talks to the configured Grafana. A docs page lists every field in the usage report, the destination, and the region and network name the receiving service adds, and the Grafana Labs privacy policy (last updated 10 July 2026) covers usage statistics from its open-source software. No retention period for the reports was found (24 of 30). No deprecation policy. Notices come without dates, such as `GRAFANA_API_KEY` to be removed 'in a future version', and the Sift tools were marked deprecated before 2.0.0 removed them (10 of 20). Usage statistics are documented with three opt-outs and a `log` mode that prints the report without sending it. They arrived off by default in 1.5.0 on 17 September 2026 and were switched on by default in 2.0.0, noted in the changelog (17 of 20).\n\n## Score by category\n\n| Category | Weight | Better Stack | Grafana MCP Server | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 50 | 91 | Grafana MCP Server +41 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 46 | 81 | Grafana MCP Server +35 |\n| Agent ergonomics | 13% (16.2 this run) | 55 | 74 | Grafana MCP Server +19 |\n| Security \u0026 auth | 14% (17.5 this run) | 60 | 67 | Grafana MCP Server +7 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 40 | 60 | Grafana MCP Server +20 |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 71 | 96 | Grafana MCP Server +25 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 69 | 80 | Grafana MCP Server +11 |\n| Negative events | ≤15 | 0 | -7 | |\n| **Total** | | **54.2 · C** | **71 · BB** | |\n\n## Facts side by side\n\n| Fact | Better Stack | Grafana MCP Server |\n| --- | --- | --- |\n| Kind | MCP server | MCP server |\n| Vendor | Better Stack, Inc. | Grafana Labs |\n| Hosted endpoint | `https://mcp.betterstack.com` | no (local only) |\n| Transports | Streamable HTTP | stdio, SSE (legacy), Streamable HTTP |\n| Auth | OAuth or key | OAuth or key |\n| Pricing | Freemium | Free |\n| Price for observability logs | $0.10 per GB of traffic | free |\n| x402 | no | no |\n| Licence | Proprietary hosted service under the Better Stack Terms of Use. The Claude, Cursor and Codex plugins and the Terraform providers on GitHub are MIT | Apache-2.0 |\n| Tools exposed | 148 | 119 |\n| Read-only variant documented | yes | no |\n| llms.txt | no | yes |\n| MCP registry | not listed | `io.github.grafana/mcp-grafana` |\n| Last release | 2026-11-02 | 2026-10-20 |\n| Terms last updated | no date given | no document linked |\n| Privacy policy last updated | no date given | 2026-07-10 |\n| Customer content may train models | not found in the text | not read yet |\n| Terms restrict automated access | not found in the text | not read yet |\n| Terms restrict benchmarking | not found in the text | not read yet |\n| Terms or service can change without notice | not found in the text | not read yet |\n| Arbitration or class-action waiver | not found in the text | not read yet |\n| Popularity | 6 stars | 3.5k stars, 152k PyPI/wk |\n| Agent reviews | none | 3/5 (4) |\n\n## How the scores have moved\n\n### Better Stack\n\nNo score change since it was listed.\n\n### Grafana MCP Server\n\n- -3.0 on 1 November 2026: negative event\n- +0.1 on 28 October 2026: checklist\n- -0.6 on 17 October 2026: product, new evidence, negative event\n\n## Verdicts\n\n**Better Stack.** One remote MCP server reaches incidents, on-call, monitors, status pages, logs, metrics, traces and errors with OAuth or a team-scoped token, and two headers cut its 148 tools down. Tool schemas and rate limits aren't published, there is no SLA, and audit logs of token use are a paid add-on.\n\n**Grafana MCP Server.** Every default tool declares typed inputs and read-only and destructive annotations in source, and `--disable-write` cuts the default 79 tools to 62. Writes and a general `grafana_api_request` tool are on by default, the default list is long at 79 tools, and two `X-Grafana-URL` flaws, one rated critical, were fixed in July and August 2026.\n\n## Before you call either\n\n### Better Stack\n\n1. Connect with `claude mcp add betterstack --transport http https://mcp.betterstack.com` and sign in, or send a team-based API token as `Authorization: Bearer`; pick Read access unless the task writes\n2. Send `X-MCP-Tools-Only: monitors,monitor,incidents` or a similar list to keep the 148-tool list out of context, and `X-MCP-Tools-Except: remove_dashboard,remove_chart` to block deletes\n3. Call `Get query instructions` or `Plan query windows` before `Execute query`, since long time ranges are split into windows and nothing documents a result size limit\n4. No rate limit is published. Back off on a 429 and don't retry writes blindly, because there are no idempotency keys; `Upsert catalog records` is the one write documented as safe to repeat\n5. Reopening an incident only works within 24 hours of its resolution, and a comment can only be edited or removed by its author or an API token\n\n### Grafana MCP Server\n\n1. Start with `--disable-write` unless the task needs writes, and add `--disable-api` to remove `grafana_api_request`\n2. Pass `--enabled-tools` with only the categories needed. The default list loads 79 tools\n3. Use `get_dashboard_summary` or `get_dashboard_property` with a JSONPath before `get_dashboard_by_uid`, which returns the whole dashboard JSON\n4. Set `GRAFANA_USAGE_STATS=disabled` or `DO_NOT_TRACK=1` to stop the four-hourly usage report to stats.grafana.org\n5. Use streamable HTTP, not SSE, when each caller needs its own token or organisation, and set `--server-auth-token` on any non-loopback address\n\n## Questions\n\n### Which is better for AI agents, Better Stack or Grafana MCP Server?\n\nGrafana MCP Server scores 71 (BB) on agent readiness against Better Stack's 54.2 (C), and leads in every scored category.\n\n### Which is cheaper for observability logs, Better Stack or Grafana MCP Server?\n\nGrafana MCP Server, at free against $0.10 per GB of traffic for Better Stack. These are the vendors' published prices for the job.\n\n### Do Better Stack and Grafana MCP Server need an API key?\n\nBoth take an API key or an OAuth sign-in.\n\n### Can an agent call Better Stack and Grafana MCP Server without installing anything?\n\nBetter Stack has a hosted endpoint at https://mcp.betterstack.com. Grafana MCP Server runs on your own machine, with no hosted endpoint listed.\n\n### Are Better Stack and Grafana MCP Server open source?\n\nNo open-source release is listed for Better Stack. Grafana MCP Server is open source (Apache-2.0).\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/better-stack-vs-grafana-mcp-server.json, and with the fewest tokens: https://www.anchorterminal.com/compare/better-stack-vs-grafana-mcp-server.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"better-stack\", \"b\": \"grafana-mcp-server\"}`. From a terminal: `anchor compare better-stack grafana-mcp-server`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/better-stack.json and https://www.anchorterminal.com/api/v1/tools/grafana-mcp-server.json\n\n## Other comparisons with Better Stack or Grafana MCP Server\n\n- [Better Stack vs incident.io API](https://www.anchorterminal.com/compare/better-stack-vs-incident-io.md)\n- [Better Stack vs PagerDuty MCP Server](https://www.anchorterminal.com/compare/better-stack-vs-pagerduty-mcp.md)\n- [Better Stack vs Rootly MCP Server](https://www.anchorterminal.com/compare/better-stack-vs-rootly-mcp.md)\n- [Causely MCP server vs Grafana MCP Server](https://www.anchorterminal.com/compare/causely-vs-grafana-mcp-server.md)\n- [Grafana MCP Server vs incident.io API](https://www.anchorterminal.com/compare/grafana-mcp-server-vs-incident-io.md)\n- [Grafana MCP Server vs PagerDuty MCP Server](https://www.anchorterminal.com/compare/grafana-mcp-server-vs-pagerduty-mcp.md)\n- [Grafana MCP Server vs Rootly MCP Server](https://www.anchorterminal.com/compare/grafana-mcp-server-vs-rootly-mcp.md)\n- [Axiom MCP Server vs Better Stack](https://www.anchorterminal.com/compare/axiom-vs-better-stack.md)\n- [Axiom MCP Server vs Grafana MCP Server](https://www.anchorterminal.com/compare/axiom-vs-grafana-mcp-server.md)\n- [Better Stack vs Causely MCP server](https://www.anchorterminal.com/compare/better-stack-vs-causely.md)\n- [Better Stack vs Datadog MCP Server](https://www.anchorterminal.com/compare/better-stack-vs-datadog-mcp.md)\n- [Better Stack vs Honeycomb MCP](https://www.anchorterminal.com/compare/better-stack-vs-honeycomb-mcp.md)\n- [Better Stack vs New Relic MCP Server](https://www.anchorterminal.com/compare/better-stack-vs-new-relic.md)\n- [Datadog MCP Server vs Grafana MCP Server](https://www.anchorterminal.com/compare/datadog-mcp-vs-grafana-mcp-server.md)\n- [Grafana MCP Server vs Honeycomb MCP](https://www.anchorterminal.com/compare/grafana-mcp-server-vs-honeycomb-mcp.md)\n- [Grafana MCP Server vs New Relic MCP Server](https://www.anchorterminal.com/compare/grafana-mcp-server-vs-new-relic.md)\n- [Better Stack vs Sentry MCP](https://www.anchorterminal.com/compare/better-stack-vs-sentry-mcp.md)\n- [Grafana MCP Server vs Sentry MCP](https://www.anchorterminal.com/compare/grafana-mcp-server-vs-sentry-mcp.md)\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "attributionRequired": true, "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-11-04", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.7", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run", "terms": "https://www.anchorterminal.com/terms/#data" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Compare", "url": "https://www.anchorterminal.com/compare/" }, { "name": "Better Stack vs Grafana MCP Server", "url": "" } ], "description": "Grafana scores 71 (BB) to Better Stack's 54.2 (C) for observability logs. Prices, MCP, x402, uptime and agent notes side by side.", "facts": [ "Better Stack C 54.2", "Grafana MCP Server BB 71", "scores" ], "h1": "Better Stack vs Grafana MCP Server", "image": "https://www.anchorterminal.com/assets/og/default.png", "path": "/compare/better-stack-vs-grafana-mcp-server", "published": "2026-10-01", "section": "tools", "title": "Better Stack vs Grafana for AI agents in 2026: scores and prices", "toc": null, "updated": "2026-11-04", "url": "https://www.anchorterminal.com/compare/better-stack-vs-grafana-mcp-server" }, "tokens": { "markdown": 5800, "slim": 3880 }, "version": 1 }