{ "data": { "a": { "slug": "honeycomb-mcp", "name": "Honeycomb MCP", "vendor": "Honeycomb", "vendorUrl": "https://www.honeycomb.io", "kind": "mcp", "category": "observability", "summary": "Honeycomb's hosted MCP server lets AI agents query a team's traces, metrics and events, run BubbleUp analysis, and create or update Boards, Triggers and SLOs. It uses OAuth or a Management API key and has US and EU endpoints.", "url": "https://www.anchorterminal.com/tools/honeycomb-mcp", "markdownUrl": "https://www.anchorterminal.com/tools/honeycomb-mcp.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/honeycomb-mcp.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/honeycomb-mcp.json", "dns": "honeycomb-mcp.grade.anchorterminal.com", "repo": "https://github.com/honeycombio/agent-skill", "license": "Proprietary hosted service under Honeycomb's Terms of Service and Honeycomb Intelligence Supplemental Terms. The companion plugin in honeycombio/agent-skill is MIT", "transports": [ "streamable-http" ], "remoteUrl": "https://mcp.honeycomb.io/mcp", "packages": [], "auth": "mixed", "authNotes": "OAuth is the documented default. The server at https://mcp.honeycomb.io/mcp points clients to its protected-resource metadata, and ui.honeycomb.io is the authorisation server, with PKCE S256 and dynamic client registration. On the consent screen the owner chooses read all, read and write all, or read or write per resource (Boards, Triggers, SLOs, recipients, Service Maps, Anomaly Signals, Canvas investigations), and every connection also gets `environments:read` and `queries:read`. Connections authorised earlier keep `mcp:read` or `mcp:write`. Unattended agents can send a Management API key as `Authorization: Bearer \u003cKEY_ID\u003e:\u003cSECRET_KEY\u003e`, with Read or Write under Model Context Protocol and access under Environments. Only a Team Owner can create that key, and its scopes can't be changed later. Access is self-serve, but a Team Owner must first enable Honeycomb Intelligence for the team.", "pricing": "byo-plan", "pricingNotes": "No separate charge. Honeycomb MCP is in the feature list of every plan, the Free plan included (20 million events and 100 million metric data points a month). Pro starts at $150 a month for 50 million events and Enterprise is quoted. The signup page asks for an email address or a Google account, and the signup guide names no payment step, so an agent's owner can start without a contract (https://www.honeycomb.io/pricing, checked 2026-10-09).", "priceSummary": "Your plan", "where": "hosted", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the MCP docs, the pricing page or the terms (checked 2026-10-09).", "endpoints": [] }, "toolCount": 36, "popularity": { "githubStars": 27, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-10-09" }, "docsUrl": "https://docs.honeycomb.io/integrations/mcp/", "llmsTxt": "https://docs.honeycomb.io/llms.txt", "registryName": "io.honeycomb/mcp", "capabilities": [ "observability.traces", "observability.metrics", "observability.logs" ], "tags": [ "official", "hosted", "mcp", "oauth", "api-key", "toolsets", "free-tier", "eu-region", "llms-txt", "mcp-registry", "status-page", "soc2" ], "lastRelease": "2026-10-09", "graded": true, "anchor": { "graded": true, "score": 60.7, "grade": "C", "agentReady": false, "rank": 579, "ranked": true, "rankOf": 1349, "categoryRank": 6, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 53, "maintenance": 81, "payments": 30, "reliability": 70, "schema": 57, "security": 68, "transparency": 70 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-04" }, "negative": 0, "verdict": "OAuth now lets an owner grant read or write per resource, so an agent can be held to queries and reads across the 36 tools, and per-tool rate limits are published. Tool input schemas are visible only after sign-in, no injection guidance was found for telemetry content, and the product changelogs disallow unnamed crawlers.", "bestFor": "Agents that investigate latency and error spikes in a team already sending traces and events to Honeycomb, and that record findings as Boards, Triggers or SLOs.", "strengths": [ "OAuth with PKCE, dynamic client registration and per-resource scopes, such as `boards:read` or `triggers:write`, chosen on the consent screen", "Per-tool rate limits are published, from 10 calls a minute for `get_service_map` to 300 for semantic-convention lookups", "Listed in the official MCP registry as `io.honeycomb/mcp` 1.0.0, under the vendor's own domain namespace", "Included in the Free plan (20 million events a month), with separate US and EU endpoints", "The Activity Log records OAuth sessions with granted scopes and each query run with its source, MCP included" ], "weaknesses": [ "Tool input schemas and annotations are visible only through `tools/list` after sign-in. The public reference gives descriptions, not parameters", "No guidance on prompt injection was found, although trace and log content returned to the agent is untrusted text", "The status page has no MCP component. US1 queries over older data failed for 67 minutes on 10 August 2026", "No SLA was found, and the AI terms allow Honeycomb to change or end any part of Honeycomb Intelligence at any time", "A Team Owner must enable Honeycomb Intelligence first, and API key scopes can't be changed after creation" ], "agentNotes": [ "Use https://mcp.honeycomb.io/mcp for US teams and https://mcp.eu1.honeycomb.io/mcp for EU teams. A team's data lives in one region only", "If no tools appear after sign-in, ask a Team Owner to enable Honeycomb Intelligence in team settings", "Ask the owner to choose Customize on the consent screen and grant only the resources the task needs. The agent sees only the tools its scopes allow", "Send `X-MCP-Toolset: query` or `canvas` to load a subset of the 36 tools. Call `get_workspace_context` first, then `find_columns`, before `run_query`", "For unattended runs send `Authorization: Bearer \u003cKEY_ID\u003e:\u003cSECRET_KEY\u003e` from a Management key with only Model Context Protocol and Environments scopes" ], "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": 60.7 } ], "editorialScores": { "ergonomics": 53, "maintenance": 81, "payments": 30, "reliability": 70, "schema": 57, "security": 68, "transparency": 64 }, "provenanceScore": 75, "checked": { "categories": { "ergonomics": "2026-11-04", "maintenance": "2026-11-04", "payments": "2026-11-04", "reliability": "2026-11-04", "schema": "2026-11-04", "security": "2026-11-04", "transparency": "2026-11-04" }, "last": "2026-11-04", "lastMoved": "2026-11-04", "oldest": "2026-11-04", "stale": false, "staleAfterDays": 90 } }, "connect": { "claudeCode": "claude mcp add honeycomb --transport http https://mcp.honeycomb.io/mcp", "config": { "mcpServers": { "honeycomb": { "url": "https://mcp.honeycomb.io/mcp" } } }, "headless": { "mcpServers": { "honeycomb": { "args": [ "-y", "mcp-remote", "https://mcp.honeycomb.io/mcp", "--header", "Authorization: Bearer $HONEYCOMB_API_KEY" ], "command": "npx", "env": { "HONEYCOMB_API_KEY": "\u003cKEY_ID\u003e:\u003cSECRET_KEY\u003e" } } } } }, "letme": { "capability": "https://letme.dev/observability.traces", "tool": "https://letme.dev/honeycomb-mcp" }, "area": "developer", "unitPrices": [ { "item": "Pro plan, 50 million events", "unit": "month", "usd": 150, "note": "Starting price. MCP is also in the Free plan at no charge" } ], "provenance": { "legalEntity": "Hound Technology, Inc. d/b/a Honeycomb", "domain": "honeycomb.io", "domainRegistered": "", "endpointOnVendorDomain": true, "terms": "https://www.honeycomb.io/terms", "privacy": "https://www.honeycomb.io/privacy", "statusPage": "https://status.honeycomb.io", "changelog": "https://changelog.honeycomb.io/", "securityTxt": "none", "checked": "2026-10-09", "notes": [ "The Terms of Service (effective 22 May 2025) name Hound Technology, Inc. d/b/a Honeycomb, are governed by California law, and incorporate the Honeycomb Intelligence Supplemental Terms at honeycomb.io/ai-terms and the Data Processing Addendum at honeycomb.io/dpa.", "The Privacy Notice (last updated 24 July 2025) covers the website and the services, and says it does not apply to content customers submit under a customer account. The DPA (last updated 22 May 2025) governs that content.", "The MCP server answers at mcp.honeycomb.io and mcp.eu1.honeycomb.io, and the OAuth server at ui.honeycomb.io, all honeycomb.io subdomains.", "www.honeycomb.io/.well-known/security.txt and /security.txt return 404. Reports go to security@honeycomb.io under a bug bounty programme whose stated scope is ui.honeycomb.io and api.honeycomb.io.", "changelog.honeycomb.io and api-changelog.honeycomb.io disallow every path to unnamed crawlers in robots.txt, so neither was read.", "The .io registry has no RDAP service that answered, so the domain's registration date is not recorded." ], "score": 75 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/honeycomb-mcp.json", "live": { "slug": "honeycomb-mcp", "probe": { "target": "https://mcp.honeycomb.io/mcp", "method": "mcp-initialize", "lastAt": "2026-11-04T21:39:27.724488683Z", "lastOk": true, "lastStatus": 401, "lastMs": 336, "lastNote": "asks for credentials", "authRequired": true, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 387, "p95ms24h": 472, "samples24h": 247, "samples30d": 6653, "days": [ { "date": "2026-10-09", "probes": 169, "ok": 169 }, { "date": "2026-10-10", "probes": 247, "ok": 247 }, { "date": "2026-10-11", "probes": 247, "ok": 247 }, { "date": "2026-10-12", "probes": 247, "ok": 247 }, { "date": "2026-10-13", "probes": 248, "ok": 248 }, { "date": "2026-10-14", "probes": 249, "ok": 249 }, { "date": "2026-10-15", "probes": 253, "ok": 253 }, { "date": "2026-10-16", "probes": 252, "ok": 252 }, { "date": "2026-10-17", "probes": 253, "ok": 253 }, { "date": "2026-10-18", "probes": 256, "ok": 256 }, { "date": "2026-10-19", "probes": 255, "ok": 255 }, { "date": "2026-10-20", "probes": 252, "ok": 252 }, { "date": "2026-10-21", "probes": 253, "ok": 253 }, { "date": "2026-10-22", "probes": 253, "ok": 253 }, { "date": "2026-10-23", "probes": 248, "ok": 248 }, { "date": "2026-10-24", "probes": 251, "ok": 251 }, { "date": "2026-10-25", "probes": 250, "ok": 250 }, { "date": "2026-10-26", "probes": 252, "ok": 252 }, { "date": "2026-10-27", "probes": 249, "ok": 249 }, { "date": "2026-10-28", "probes": 251, "ok": 251 }, { "date": "2026-10-29", "probes": 248, "ok": 248 }, { "date": "2026-10-30", "probes": 249, "ok": 249 }, { "date": "2026-10-31", "probes": 251, "ok": 251 }, { "date": "2026-11-01", "probes": 249, "ok": 249 }, { "date": "2026-11-02", "probes": 250, "ok": 250 }, { "date": "2026-11-03", "probes": 248, "ok": 248 }, { "date": "2026-11-04", "probes": 223, "ok": 223 } ] }, "vendorStatus": { "page": "https://status.honeycomb.io", "indicator": "none", "summary": "All Systems Operational", "checkedAt": "2026-11-04T21:38:19.832596024Z" }, "versions": [ { "registry": "mcp-registry", "name": "io.honeycomb/mcp", "version": "1.0.0", "seenAt": "2026-11-04T17:11:44.552009456Z" } ], "githubStars": 28, "securityTxt": { "url": "https://honeycomb.io/.well-known/security.txt", "state": "none", "checkedAt": "2026-11-04T15:52:49.534670121Z" }, "llmsTxt": { "url": "https://docs.honeycomb.io/llms.txt", "ok": true, "status": 200, "checkedAt": "2026-11-04T15:23:11.617931168Z" }, "domain": { "domain": "honeycomb.io", "checkedAt": "2026-11-01T14:12:37.895563706Z" }, "pages": [ { "url": "https://changelog.honeycomb.io/", "kind": "changelog", "status": 0, "checkedAt": "2026-11-04T08:16:22.238769817Z", "changedAt": "0001-01-01T00:00:00Z", "blockedByRobots": true }, { "url": "https://www.honeycomb.io/pricing", "kind": "pricing", "status": 200, "checkedAt": "2026-11-04T08:41:02.167425031Z", "changedAt": "2026-11-03T08:06:55.29202553Z", "fingerprint": "cc55e59dbc8d" }, { "url": "https://www.honeycomb.io/privacy", "kind": "privacy", "status": 200, "checkedAt": "2026-11-04T08:41:04.630185931Z", "changedAt": "2026-11-03T08:06:57.784332085Z", "fingerprint": "51602bc012d0" }, { "url": "https://www.honeycomb.io/terms", "kind": "terms", "status": 200, "checkedAt": "2026-11-04T08:41:06.760623981Z", "changedAt": "2026-11-03T08:07:00.196017318Z", "fingerprint": "b691d7c4af5c" } ], "mcpTools": { "url": "https://mcp.honeycomb.io/mcp", "checkedAt": "2026-11-04T01:22:47.870034582Z", "status": "auth", "note": "asks for credentials before listing its tools", "changedAt": "2026-10-09T21:40:43.761722159Z" }, "github": { "repo": "honeycombio/agent-skill", "checkedAt": "2026-11-02T08:54:48.0222443Z", "stars": 28, "archived": false, "issuesEnabled": true, "pushedAt": "2026-10-09T22:47:47Z", "openIssues": 3, "openPRs": 1, "issuesOpened90d": 1, "issuesClosed90d": 1, "prsMerged90d": 4, "ci": { "branch": "main", "runs": 50, "success": 43, "failure": 7, "other": 0, "latest": "success", "latestAt": "2026-10-09T22:48:09Z" }, "replies": { "sampled": 9, "unanswered": 3, "medianDays": 3.7 } }, "updatedAt": "2026-11-04T21:39:27.724488683Z" } }, "answer": "New Relic MCP Server and Honeycomb MCP score within a point of each other on agent readiness, 61 (C) and 60.7 (C). Honeycomb MCP leads on reliability and schema \u0026 documentation.", "b": { "slug": "new-relic", "name": "New Relic MCP Server", "vendor": "New Relic, Inc.", "vendorUrl": "https://newrelic.com", "kind": "mcp", "category": "observability", "summary": "New Relic's hosted MCP server at mcp.newrelic.com gives AI agents read access to an account's entities, NRQL data, alerts, issues, logs, errors, traces and deployments, with OAuth or a user API key, on top of the NerdGraph API.", "url": "https://www.anchorterminal.com/tools/new-relic", "markdownUrl": "https://www.anchorterminal.com/tools/new-relic.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/new-relic.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/new-relic.json", "dns": "new-relic.grade.anchorterminal.com", "repo": "https://github.com/newrelic/nr-skills-hub", "transports": [ "streamable-http" ], "remoteUrl": "https://mcp.newrelic.com/mcp/", "packages": [], "auth": "oauth", "authNotes": "OAuth 2.0 (recommended) with scopes `observability:read`, `offline` and `offline_access`; access tokens expire after 5 minutes and are refreshed, with sign-in again after 24 hours. Or a user API key (`NRAK-...`) in an `api-key` header, which the docs say only works where stated and excludes the AI analysis, gap and preview tools. An administrator must leave the MCP toggles on in feature control manager and the user needs an organisation-scoped read role. EU and JP accounts use mcp.eu.newrelic.com and mcp.jp.newrelic.com.", "pricing": "freemium", "pricingNotes": "MCP is available on every New Relic account, including the free tier (100 GB of ingest a month, one full platform user, no card). Most tools are metered as Core Compute Units against the account's allocation and four AI analysis tools as Advanced Compute Units; no per-call price is published. Data beyond 100 GB costs $0.40 per GB; user and edition prices are quoted on request (https://newrelic.com/pricing).", "priceSummary": "$0.40 / GB", "where": "hosted", "x402": { "level": "no", "evidence": "No x402, MPP or L402 in the MCP docs, pricing page or terms (checked 2026-11-03). MCP calls are metered as compute units on the account's bill.", "endpoints": [] }, "toolCount": 40, "popularity": { "githubStars": null, "npmWeekly": null, "pypiWeekly": null, "asOf": "2026-11-03" }, "docsUrl": "https://docs.newrelic.com/docs/agentic-ai/mcp/overview/", "llmsTxt": "https://docs.newrelic.com/llms.txt", "registryName": "com.newrelic/mcp-server", "capabilities": [ "observability.metrics", "observability.logs", "observability.errors", "observability.traces", "observability.incidents" ], "tags": [ "hosted", "freemium", "free-tier", "no-card", "mcp", "remote-mcp", "mcp-registry", "oauth", "read-only-mode", "llms-txt", "status-page", "security-txt", "bug-bounty", "soc2", "eu-region", "closed-source" ], "lastRelease": "2026-10-28", "graded": true, "anchor": { "graded": true, "score": 61, "grade": "C", "agentReady": false, "rank": 567, "ranked": true, "rankOf": 1349, "categoryRank": 5, "methodology": "0.7", "run": "2026-10-01", "scores": { "ergonomics": 59, "maintenance": 80, "payments": 30, "reliability": 60, "schema": 51, "security": 76, "transparency": 81 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-11-03" }, "negative": 0, "verdict": "OAuth with a read scope, 5-minute access tokens, role-bound calls and a queryable audit trail make this a safe read surface over an account's data, and the free tier includes it. 40 tools need tag filtering, the schemas aren't public, there is no per-call price, and no dated GA notice was found.", "bestFor": "Agents investigating incidents, errors, deployments and performance in an existing New Relic account, where reads are enough.", "strengths": [ "OAuth 2.0 with `observability:read`, access tokens that expire after 5 minutes, revocation by user or organisation toggle, and every call bound to the user's roles", "`include-tags` header limits the 40 tools to the groups a task needs, such as `discovery,alerting`", "Sign-ins and NRQL sent through the server are queryable in NrAuditEvent and NrdbQuery with a product label", "Available on the free tier (100 GB a month, no card) and rate-limited at a published 2,000 tool calls an hour per user", "llms.txt and Markdown twins of every docs page, a dated what's-new feed and a verified registry namespace" ], "weaknesses": [ "Tool input schemas are visible only through an authenticated session; the docs give one-line descriptions and no parameters", "40 tools in the default list, above the 30-tool line, unless the client sends `include-tags`", "No per-call price; tools are metered as Core and Advanced Compute Units on the account's bill", "Public preview since 4 November 2025 with no dated general availability notice found, and preview tools mixed into the list", "Logs, errors and traces come back as untrusted text with no prompt-injection guidance; the trust centre renders through a script our reader can't run" ], "agentNotes": [ "Connect with OAuth; an `api-key` header reaches only the core tools and the docs say the API key route works only where stated", "Send `include-tags` with the groups the task needs (for example `discovery,data-access`) to keep the tool list under 30", "Call `list_available_new_relic_accounts` first and pass the account id to every query; `convert_time_period_to_epoch_ms` turns phrases such as last 30 minutes into the epoch bounds the other tools take", "Stay under 2,000 tool calls an hour per user and 25 concurrent NerdGraph requests; a 429 clears as in-flight requests finish", "Ask the organisation administrator to turn on the Preview Tools, API Keys, Atlassian or AWS toggles in feature control manager before relying on those tools; they are off by default" ], "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": 61 } ], "editorialScores": { "ergonomics": 59, "maintenance": 80, "payments": 30, "reliability": 60, "schema": 51, "security": 76, "transparency": 63 }, "provenanceScore": 98, "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": { "claudeCode": "claude mcp add --transport http --scope user newrelic https://mcp.newrelic.com/mcp/", "config": { "mcpServers": { "newrelic": { "type": "http", "url": "https://mcp.newrelic.com/mcp/" } } } }, "letme": { "capability": "https://letme.dev/observability.metrics", "tool": "https://letme.dev/new-relic" }, "area": "developer", "unitPrices": [ { "item": "Data ingest beyond 100 GB a month", "unit": "gb", "usd": 0.4, "note": "100 GB a month free on every edition. User and compute prices are quoted on request" } ], "provenance": { "legalEntity": "New Relic, Inc.", "domain": "newrelic.com", "domainRegistered": "2006-04-19", "endpointOnVendorDomain": true, "terms": "https://newrelic.com/termsandconditions/paid", "privacy": "https://newrelic.com/termsandconditions/privacy", "statusPage": "https://status.newrelic.com", "changelog": "https://docs.newrelic.com/whats-new/", "securityTxt": "valid", "checked": "2026-11-03", "notes": [ "The paid terms of service are dated 8 November 2022 and the unpaid terms 21 October 2021; the master cloud agreement path is disallowed by newrelic.com/robots.txt and wasn't read. The privacy notice is dated 14 September 2026.", "newrelic.com/robots.txt disallows /signup/, /lp/, the master cloud agreement and a few files; /security redirected to trust.newrelic.com, which renders through a script and was not readable. docs.newrelic.com allows everything.", "security.txt at newrelic.com/.well-known/security.txt points to HackerOne for contact and policy.", "status.newrelic.com is an Atlassian Statuspage with 115 components across US, EU and JP; the history page needs JavaScript, so the 90-day record was read from history.rss, which holds 25 incidents back to 12 January 2026.", "RDAP gives newrelic.com a registration date of 19 April 2006. firstReleased is 4 November 2025, the public preview date of the MCP server." ], "score": 98 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/new-relic.json", "live": { "slug": "new-relic", "probe": { "target": "https://mcp.newrelic.com/mcp/", "method": "mcp-initialize", "lastAt": "2026-11-04T21:39:36.212939859Z", "lastOk": true, "lastStatus": 401, "lastMs": 302, "lastNote": "asks for credentials", "authRequired": true, "uptime24h": 100, "uptime30d": 100, "p50ms24h": 327, "p95ms24h": 416, "samples24h": 64, "samples30d": 64, "days": [ { "date": "2026-11-04", "probes": 64, "ok": 64 } ] }, "vendorStatus": { "page": "https://status.newrelic.com", "indicator": "none", "summary": "All Systems Operational", "checkedAt": "2026-11-04T21:38:31.653792842Z" }, "versions": [ { "registry": "mcp-registry", "name": "com.newrelic/mcp-server", "version": "0.1.0", "seenAt": "2026-11-04T17:11:44.552009456Z" } ], "githubStars": 7, "securityTxt": { "url": "https://newrelic.com/.well-known/security.txt", "state": "valid", "checkedAt": "2026-11-04T15:54:45.364182842Z" }, "updatedAt": "2026-11-04T21:39:36.212939859Z" } }, "edges": [ { "by": 11, "category": "transparency", "leads": "new-relic", "name": "Transparency \u0026 trust", "reasons": { "a": "The hosted server is closed, under published Terms of Service and Honeycomb Intelligence Supplemental Terms. The companion plugin is MIT (17). The privacy notice, the DPA and an AI policy page dated 21 September 2026 are public. The AI page says MCP's model provider is whichever the client uses, and that Honeycomb uses no provider that trains foundation models on input. Telemetry is deleted after the retention window, 60 days for most customers. The terms also let Honeycomb use customer data to improve its services, and the AI page says Honeycomb may fine-tune pre-trained models (22). The self-hosted server carries a deprecation notice dated 21 August 2025. No deprecation policy was found, the AI terms allow any part of Honeycomb Intelligence to be changed or ended at any time with commercially reasonable efforts at notice, and the docs say rate limits may change (6). The subprocessor list is public with purposes and countries, the DPA promises 30 days' notice of new subprocessors, and customers choose a US or EU data location (19).", "b": "Closed service with clear dated terms (paid 8 November 2022, unpaid 21 October 2021) naming New Relic, Inc. (15). A privacy notice of 14 September 2026, a data retention page with per-namespace periods, a security policy, and the MCP docs' statement that gap-analysis responses aren't stored; the master cloud agreement and the trust centre couldn't be read (22). End-of-life notices with dates (REST API keys, 2025) and a pre-release policy that says preview services can change or end (16). US, EU and JP regions are documented for the server and the platform, but no sub-processor list was found on a readable page (10). 15 + 22 + 16 + 10 = 63." }, "scores": { "a": 70, "b": 81 } }, { "by": 10, "category": "reliability", "leads": "honeycomb-mcp", "name": "Reliability", "reasons": { "a": "Scored as a hosted MCP server. status.honeycomb.io is an Atlassian Statuspage with components for US1, EU1, the website and the sandbox. It has no MCP component, though MCP incidents are posted there, which takes 15 as sentry-mcp and slack-mcp take for a vendor page with no MCP component (15). history.atom lists five incidents from 6 August to 4 November 2026, each limited to one region, one product or a subset of requests. US1 queries over older data erroring for 67 minutes on 10 August, Canvas not loading for 38 minutes on 21 August, degraded US event ingest for 12 minutes on 20 October, slower query execution and SLO processing in the US from 07:04 to 09:26 PDT on 20 October, and SLO, Trigger and anomaly emails failing for 15 minutes on 22 October. None took the query path down for everyone, so minor only (20). Per-tool limits are published with numbers (15). Rate-limit errors carry the time to retry after, which counts as Retry-After alone. No backoff guidance or idempotency keys for the nine write tools were found (10). No SLA found in the terms, and honeycomb.io/sla returned 404 on 9 October (0). The docs still list MCP among betas graduated to general availability (10). 15+20+15+10+0+10 = 70.", "b": "Scored as a hosted MCP server. status.newrelic.com is an Atlassian Statuspage with 115 components across the US, EU and JP regions and no MCP component (20). history.rss holds seven incidents in the 90 days to 3 November 2026, among them delayed queries and alerts in the US for 49 minutes on 2 November, synthetics ping delays the same day, the JP entity platform unavailable for 67 minutes on 28 October, delayed US data and alerts from 19:24 to 22:04 UTC on 25 September, intermittent UI errors for two and a half hours the same night, a Salesforce support outage on 16 September and a 10-minute EU metric issue. Two regional incidents over an hour on the query and alert paths the server depends on, so one major (10). Published numbers, 2,000 tool calls an hour per user on the MCP server and 25 concurrent NerdGraph requests per user (15). NerdGraph answers 429 on concurrency with guidance to batch or pool, but no Retry-After is documented and nothing covers retries of MCP calls; the tools are reads, so idempotency doesn't arise (8). The pricing page lists support response SLAs, and the paid terms promise commercially reasonable efforts, so no availability SLA (0). Public preview on 4 November 2025, available to all accounts on 3 November 2026 with preview tools marked separately, but no dated GA notice was found (7). 20 + 10 + 15 + 8 + 0 + 7 = 60." }, "scores": { "a": 70, "b": 60 } }, { "by": 8, "category": "security", "leads": "new-relic", "name": "Security \u0026 auth", "reasons": { "a": "OAuth authorisation code with PKCE S256, dynamic client registration, client ID metadata documents and refresh tokens. Since the 9 October check the consent screen lets the owner choose read all, read and write all, or a custom choice per resource, with scopes such as `boards:read`, `triggers:write` and `investigations:write` beside `environments:read` and `queries:read`. Connections made earlier keep `mcp:read` or `mcp:write`. The alternative is a Management API key whose scopes are fixed at creation and which can be disabled or deleted (28). Tools are read-only unless a write scope is granted, members with a read-only role can choose only read scopes, API keys are limited to chosen environments, and there is no delete tool. No confirmation step for writes was found (15). Trace, log and span content is untrusted text, and no injection guidance was found in the MCP docs (2). The Activity Log records OAuth sessions with requested and granted scopes and one event per query run with its source, MCP included. The compliance page describes activity logging as for enterprise customers (11). SOC 2 Type II each year with penetration tests by an independent firm, HIPAA BAAs and an email disclosure route under a bug bounty page whose rewards are all at Honeycomb's discretion, so the bounty isn't counted. No security.txt, no stated acknowledgement time, and the scope names ui.honeycomb.io and api.honeycomb.io, not the MCP hosts. ISO 27001 is claimed only for the cloud hosts. One certification and a disclosure route, set between mailersend's 8 (disclosure route, discretionary bounty, no certification) and daytona's 15 (disclosure route, SOC 2 and ISO 27001, discretionary reward not counted) (12).", "b": "OAuth 2.0 with the scopes `observability:read`, `offline` and `offline_access`, access tokens that expire after 5 minutes, refresh for 24 hours, revocation by deleting the user or turning off the organisation toggle, and the token acting as the user with no more than their roles; the alternative is a user API key in a header (28). A read scope, organisation toggles that default the API key, preview, Atlassian and AWS routes to off, an organisation-scoped role requirement, and no write tools in the list (16). Logs, error messages and trace attributes are untrusted text and no injection guidance was found (2). NrAuditEvent records sign-ins with IP addresses and NrdbQuery labels queries sent through the server, both queryable by the operator (13). security.txt pointing to a HackerOne programme, a published security policy citing SOC 2 Type 2 and ISO 27001 audits of its facilities and its own penetration testing, and an ISO 42001 certification announced on 29 September 2026; the trust centre itself renders through a script and was unread (17). 28 + 16 + 2 + 13 + 17 = 76." }, "scores": { "a": 68, "b": 76 } }, { "by": 6, "category": "schema", "leads": "honeycomb-mcp", "name": "Schema \u0026 documentation", "reasons": { "a": "The server is closed source and `tools/list` needs sign-in, so we could not read the input schemas. Partial credit for a contract that exists but is not public (12). docs.honeycomb.io publishes llms.txt and a Markdown twin of every page (10). The tools reference says what each of the 36 tools returns and when an agent reaches for it, for example calling `get_workspace_context` first and `list_memories` before `read_memory`. These are docs, not the definitions the model sees (14). Parameters are mentioned only in passing, with no types, enums or required fields published (4). A use-cases page and a troubleshooting page document the Unauthorized, rate-limit, permission and not-found errors with causes and fixes (9). The registry entry carries version 1.0.0 and Honeycomb links a UI changelog and an API changelog, but changelog.honeycomb.io still disallows unnamed crawlers, so we could not confirm MCP entries (8).", "b": "Tool input schemas are only visible through an authenticated session; the tool reference names 40 tools with one-line descriptions, so partial credit for a contract that exists but isn't public (10). docs.newrelic.com publishes llms.txt, updated 3 November 2026, and a Markdown twin of every page (10). Each tool's line says what it does, tags say which group it belongs to, and the reference says which tools need OAuth and which are preview; nothing says when not to use a tool (12). No parameters, types or enums are published; the setup examples show only the `include-tags` header (4). Example natural-language queries per group, a troubleshooting page for OAuth, API key, permission and connection errors, and no error code list (7). A dated what's-new feed and a 0.1.0 registry version, but no MCP changelog or server version in the docs (8). 10 + 10 + 12 + 4 + 7 + 8 = 51." }, "scores": { "a": 57, "b": 51 } }, { "by": 6, "category": "ergonomics", "leads": "new-relic", "name": "Agent ergonomics", "reasons": { "a": "36 tools is above the 30 line (5). The `X-MCP-Toolset` header loads `query` or `canvas` subsets, and the docs don't say how many tools each holds (add 8, so 13). Size controls are described per tool, such as up to 100 datasets or 100 columns a call, subtree zoom on `get_trace` and query limits. Pagination parameters aren't documented (10). Errors are documented with causes and fixes, and the rate-limit message names the retry time. No error code list was found (12). The reference labels each tool read or write and `update_slo` keeps omitted fields. No idempotency keys were found, and we couldn't confirm readOnlyHint or destructiveHint annotations (8). `get_workspace_context` takes no parameters, relative time is the default, and setup is one command or link in nine documented clients. SDKs don't apply (10).", "b": "40 tools in the default list, above the 30-tool line (5), with the `include-tags` header to load only named groups (add 10, so 15). Time windows on most tools, `convert_time_period_to_epoch_ms`, filters on incidents, errors and logs, and NRQL's own LIMIT and FACET; no documented page size (14). The troubleshooting page covers setup errors with fixes, and the docs say nothing about tool-level error bodies (8). Every tool is a read (query, list, fetch, analyse) and the gap tool reads files the client sends, so a retry is safe; no readOnlyHint or destructiveHint was confirmed because tools/list wasn't read (12). Natural-language prompts need no parameters from the user, and the API behind the server has official NerdGraph clients; nothing agent-specific beyond skills (10). 15 + 14 + 8 + 12 + 10 = 59." }, "scores": { "a": 53, "b": 59 } } ], "facts": [ { "a": "MCP server", "b": "MCP server", "name": "Kind" }, { "a": "Honeycomb", "b": "New Relic, Inc.", "name": "Vendor" }, { "a": "https://mcp.honeycomb.io/mcp", "b": "https://mcp.newrelic.com/mcp/", "name": "Hosted endpoint" }, { "a": "Streamable HTTP", "b": "Streamable HTTP", "name": "Transports" }, { "a": "OAuth or key", "b": "OAuth", "name": "Auth" }, { "a": "Your plan", "b": "Freemium", "name": "Pricing" }, { "a": "no", "b": "no", "name": "x402" }, { "a": "Proprietary hosted service under Honeycomb's Terms of Service and Honeycomb Intelligence Supplemental Terms. The companion plugin in honeycombio/agent-skill is MIT", "b": "none", "name": "Licence" }, { "a": "36", "b": "40", "name": "Tools exposed" }, { "a": "no", "b": "yes", "name": "Read-only variant documented" }, { "a": "yes", "b": "yes", "name": "llms.txt" }, { "a": "io.honeycomb/mcp", "b": "com.newrelic/mcp-server", "name": "MCP registry" }, { "a": "2026-10-09", "b": "2026-10-28", "name": "Last release" }, { "a": "2025-05-22", "b": "2022-11-08", "name": "Terms last updated" }, { "a": "no date given", "b": "2026-09-14", "name": "Privacy policy last updated" }, { "a": "not found in the text", "b": "not found in the text", "name": "Customer content may train models" }, { "a": "not found in the text", "b": "not found in the text", "name": "Terms restrict automated access" }, { "a": "not found in the text", "b": "yes", "name": "Terms restrict benchmarking" }, { "a": "yes", "b": "not found in the text", "name": "Terms or service can change without notice" }, { "a": "not found in the text", "b": "not found in the text", "name": "Arbitration or class-action waiver" }, { "a": "27 stars", "b": "none", "name": "Popularity" } ], "faq": [ { "answer": "New Relic MCP Server and Honeycomb MCP score within a point of each other on agent readiness, 61 (C) and 60.7 (C). Honeycomb MCP leads on reliability and schema \u0026 documentation.", "question": "Which is better for AI agents, Honeycomb MCP or New Relic MCP Server?" }, { "answer": "Honeycomb MCP takes an API key or an OAuth sign-in. New Relic MCP Server uses an OAuth sign-in.", "question": "Do Honeycomb MCP and New Relic MCP Server need an API key?" }, { "answer": "Yes. Honeycomb MCP has a hosted endpoint at https://mcp.honeycomb.io/mcp and New Relic MCP Server at https://mcp.newrelic.com/mcp/.", "question": "Can an agent call Honeycomb MCP and New Relic MCP Server without installing anything?" } ], "goodFor": [ { "aheadOn": [ "Reliability, 70 against 60", "Schema \u0026 documentation, 57 against 51" ], "also": null, "goodFor": "Agents that investigate latency and error spikes in a team already sending traces and events to Honeycomb, and that record findings as Boards, Triggers or SLOs.", "slug": "honeycomb-mcp", "watchFor": "Tool input schemas and annotations are visible only through `tools/list` after sign-in. The public reference gives descriptions, not parameters" }, { "aheadOn": [ "Agent ergonomics, 59 against 53", "Security \u0026 auth, 76 against 68", "Transparency \u0026 trust, 81 against 70" ], "also": [ "Free to start without a card" ], "goodFor": "Agents investigating incidents, errors, deployments and performance in an existing New Relic account, where reads are enough.", "slug": "new-relic", "watchFor": "Tool input schemas are visible only through an authenticated session; the docs give one-line descriptions and no parameters" } ], "history": { "honeycomb-mcp": [ { "causes": [ "product", "correction" ], "date": "2026-11-04", "delta": 2.2, "from": 58.5, "fromGrade": "C", "to": 60.7, "toGrade": "C" }, { "causes": [ "method" ], "date": "2026-10-27", "delta": 0.2, "from": 58.3, "fromGrade": "C", "to": 58.5, "toGrade": "C" } ], "new-relic": null }, "job": { "capability": "observability.traces", "name": "Observability traces" }, "others": [ { "json": "https://www.anchorterminal.com/compare/causely-vs-new-relic.json", "title": "Causely MCP server vs New Relic MCP Server", "url": "https://www.anchorterminal.com/compare/causely-vs-new-relic" }, { "json": "https://www.anchorterminal.com/compare/incident-io-vs-new-relic.json", "title": "incident.io API vs New Relic MCP Server", "url": "https://www.anchorterminal.com/compare/incident-io-vs-new-relic" }, { "json": "https://www.anchorterminal.com/compare/new-relic-vs-pagerduty-mcp.json", "title": "New Relic MCP Server vs PagerDuty MCP Server", "url": "https://www.anchorterminal.com/compare/new-relic-vs-pagerduty-mcp" }, { "json": "https://www.anchorterminal.com/compare/new-relic-vs-rootly-mcp.json", "title": "New Relic MCP Server vs Rootly MCP Server", "url": "https://www.anchorterminal.com/compare/new-relic-vs-rootly-mcp" }, { "json": "https://www.anchorterminal.com/compare/axiom-vs-honeycomb-mcp.json", "title": "Axiom MCP Server vs Honeycomb MCP", "url": "https://www.anchorterminal.com/compare/axiom-vs-honeycomb-mcp" }, { "json": "https://www.anchorterminal.com/compare/axiom-vs-new-relic.json", "title": "Axiom MCP Server vs New Relic MCP Server", "url": "https://www.anchorterminal.com/compare/axiom-vs-new-relic" }, { "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/causely-vs-honeycomb-mcp.json", "title": "Causely MCP server vs Honeycomb MCP", "url": "https://www.anchorterminal.com/compare/causely-vs-honeycomb-mcp" }, { "json": "https://www.anchorterminal.com/compare/datadog-mcp-vs-honeycomb-mcp.json", "title": "Datadog MCP Server vs Honeycomb MCP", "url": "https://www.anchorterminal.com/compare/datadog-mcp-vs-honeycomb-mcp" }, { "json": "https://www.anchorterminal.com/compare/datadog-mcp-vs-new-relic.json", "title": "Datadog MCP Server vs New Relic MCP Server", "url": "https://www.anchorterminal.com/compare/datadog-mcp-vs-new-relic" }, { "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/new-relic-vs-sentry-mcp.json", "title": "New Relic MCP Server vs Sentry MCP", "url": "https://www.anchorterminal.com/compare/new-relic-vs-sentry-mcp" }, { "json": "https://www.anchorterminal.com/compare/honeycomb-mcp-vs-sentry-mcp.json", "title": "Honeycomb MCP vs Sentry MCP", "url": "https://www.anchorterminal.com/compare/honeycomb-mcp-vs-sentry-mcp" } ], "panel": null, "scores": [ { "by": 10, "edge": "honeycomb-mcp", "honeycomb-mcp": 70, "key": "reliability", "name": "Reliability", "new-relic": 60, "weight": 16 }, { "key": "performance", "name": "Performance", "pending": true, "weight": 10 }, { "by": 6, "edge": "honeycomb-mcp", "honeycomb-mcp": 57, "key": "schema", "name": "Schema \u0026 documentation", "new-relic": 51, "weight": 13 }, { "by": 6, "edge": "new-relic", "honeycomb-mcp": 53, "key": "ergonomics", "name": "Agent ergonomics", "new-relic": 59, "weight": 13 }, { "by": 8, "edge": "new-relic", "honeycomb-mcp": 68, "key": "security", "name": "Security \u0026 auth", "new-relic": 76, "weight": 14 }, { "by": 0, "edge": "", "honeycomb-mcp": 30, "key": "payments", "name": "Payments \u0026 pricing", "new-relic": 30, "weight": 10 }, { "key": "tasks", "name": "Task success", "pending": true, "weight": 10 }, { "by": 1, "edge": "honeycomb-mcp", "honeycomb-mcp": 81, "key": "maintenance", "name": "Maintenance \u0026 community", "new-relic": 80, "weight": 7 }, { "by": 11, "edge": "new-relic", "honeycomb-mcp": 70, "key": "transparency", "name": "Transparency \u0026 trust", "new-relic": 81, "weight": 7 } ], "summary": "New Relic MCP Server and Honeycomb MCP score within a point of each other on agent readiness, 61 (C) and 60.7 (C). Honeycomb MCP leads on reliability and schema \u0026 documentation. Both do observability traces.", "tasks": null, "tutorials": [], "verdicts": { "honeycomb-mcp": "OAuth now lets an owner grant read or write per resource, so an agent can be held to queries and reads across the 36 tools, and per-tool rate limits are published. Tool input schemas are visible only after sign-in, no injection guidance was found for telemetry content, and the product changelogs disallow unnamed crawlers.", "new-relic": "OAuth with a read scope, 5-minute access tokens, role-bound calls and a queryable audit trail make this a safe read surface over an account's data, and the free tier includes it. 40 tools need tag filtering, the schemas aren't public, there is no per-call price, and no dated GA notice was found." } }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/honeycomb-mcp-vs-new-relic", "json": "https://www.anchorterminal.com/compare/honeycomb-mcp-vs-new-relic.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/honeycomb-mcp-vs-new-relic.md", "mcp": "https://www.anchorterminal.com/mcp", "slim": "https://www.anchorterminal.com/compare/honeycomb-mcp-vs-new-relic.min.md" }, "markdown": "New Relic MCP Server and Honeycomb MCP score within a point of each other on agent readiness, 61 (C) and 60.7 (C). Honeycomb MCP leads on reliability and schema \u0026 documentation. Both do observability traces.\n\n- Honeycomb MCP: grade C, 60.7/100, rank #579 of 1349. Markdown https://www.anchorterminal.com/tools/honeycomb-mcp.md · JSON https://www.anchorterminal.com/api/v1/tools/honeycomb-mcp.json\n- New Relic MCP Server: grade C, 61/100, rank #567 of 1349. Markdown https://www.anchorterminal.com/tools/new-relic.md · JSON https://www.anchorterminal.com/api/v1/tools/new-relic.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### Honeycomb MCP (C)\n\nGood for: Agents that investigate latency and error spikes in a team already sending traces and events to Honeycomb, and that record findings as Boards, Triggers or SLOs.\n\nAhead on:\n- Reliability, 70 against 60\n- Schema \u0026 documentation, 57 against 51\n\nWatch for: Tool input schemas and annotations are visible only through `tools/list` after sign-in. The public reference gives descriptions, not parameters\n\n### New Relic MCP Server (C)\n\nGood for: Agents investigating incidents, errors, deployments and performance in an existing New Relic account, where reads are enough.\n\nAhead on:\n- Agent ergonomics, 59 against 53\n- Security \u0026 auth, 76 against 68\n- Transparency \u0026 trust, 81 against 70\n\nAlso in its favour:\n- Free to start without a card\n\nWatch for: Tool input schemas are visible only through an authenticated session; the docs give one-line descriptions and no parameters\n\n\n## Why the edge\n\n### Transparency \u0026 trust: New Relic MCP Server +11\n\n- Honeycomb MCP (70): The hosted server is closed, under published Terms of Service and Honeycomb Intelligence Supplemental Terms. The companion plugin is MIT (17). The privacy notice, the DPA and an AI policy page dated 21 September 2026 are public. The AI page says MCP's model provider is whichever the client uses, and that Honeycomb uses no provider that trains foundation models on input. Telemetry is deleted after the retention window, 60 days for most customers. The terms also let Honeycomb use customer data to improve its services, and the AI page says Honeycomb may fine-tune pre-trained models (22). The self-hosted server carries a deprecation notice dated 21 August 2025. No deprecation policy was found, the AI terms allow any part of Honeycomb Intelligence to be changed or ended at any time with commercially reasonable efforts at notice, and the docs say rate limits may change (6). The subprocessor list is public with purposes and countries, the DPA promises 30 days' notice of new subprocessors, and customers choose a US or EU data location (19).\n- New Relic MCP Server (81): Closed service with clear dated terms (paid 8 November 2022, unpaid 21 October 2021) naming New Relic, Inc. (15). A privacy notice of 14 September 2026, a data retention page with per-namespace periods, a security policy, and the MCP docs' statement that gap-analysis responses aren't stored; the master cloud agreement and the trust centre couldn't be read (22). End-of-life notices with dates (REST API keys, 2025) and a pre-release policy that says preview services can change or end (16). US, EU and JP regions are documented for the server and the platform, but no sub-processor list was found on a readable page (10). 15 + 22 + 16 + 10 = 63.\n\n### Reliability: Honeycomb MCP +10\n\n- Honeycomb MCP (70): Scored as a hosted MCP server. status.honeycomb.io is an Atlassian Statuspage with components for US1, EU1, the website and the sandbox. It has no MCP component, though MCP incidents are posted there, which takes 15 as sentry-mcp and slack-mcp take for a vendor page with no MCP component (15). history.atom lists five incidents from 6 August to 4 November 2026, each limited to one region, one product or a subset of requests. US1 queries over older data erroring for 67 minutes on 10 August, Canvas not loading for 38 minutes on 21 August, degraded US event ingest for 12 minutes on 20 October, slower query execution and SLO processing in the US from 07:04 to 09:26 PDT on 20 October, and SLO, Trigger and anomaly emails failing for 15 minutes on 22 October. None took the query path down for everyone, so minor only (20). Per-tool limits are published with numbers (15). Rate-limit errors carry the time to retry after, which counts as Retry-After alone. No backoff guidance or idempotency keys for the nine write tools were found (10). No SLA found in the terms, and honeycomb.io/sla returned 404 on 9 October (0). The docs still list MCP among betas graduated to general availability (10). 15+20+15+10+0+10 = 70.\n- New Relic MCP Server (60): Scored as a hosted MCP server. status.newrelic.com is an Atlassian Statuspage with 115 components across the US, EU and JP regions and no MCP component (20). history.rss holds seven incidents in the 90 days to 3 November 2026, among them delayed queries and alerts in the US for 49 minutes on 2 November, synthetics ping delays the same day, the JP entity platform unavailable for 67 minutes on 28 October, delayed US data and alerts from 19:24 to 22:04 UTC on 25 September, intermittent UI errors for two and a half hours the same night, a Salesforce support outage on 16 September and a 10-minute EU metric issue. Two regional incidents over an hour on the query and alert paths the server depends on, so one major (10). Published numbers, 2,000 tool calls an hour per user on the MCP server and 25 concurrent NerdGraph requests per user (15). NerdGraph answers 429 on concurrency with guidance to batch or pool, but no Retry-After is documented and nothing covers retries of MCP calls; the tools are reads, so idempotency doesn't arise (8). The pricing page lists support response SLAs, and the paid terms promise commercially reasonable efforts, so no availability SLA (0). Public preview on 4 November 2025, available to all accounts on 3 November 2026 with preview tools marked separately, but no dated GA notice was found (7). 20 + 10 + 15 + 8 + 0 + 7 = 60.\n\n### Security \u0026 auth: New Relic MCP Server +8\n\n- Honeycomb MCP (68): OAuth authorisation code with PKCE S256, dynamic client registration, client ID metadata documents and refresh tokens. Since the 9 October check the consent screen lets the owner choose read all, read and write all, or a custom choice per resource, with scopes such as `boards:read`, `triggers:write` and `investigations:write` beside `environments:read` and `queries:read`. Connections made earlier keep `mcp:read` or `mcp:write`. The alternative is a Management API key whose scopes are fixed at creation and which can be disabled or deleted (28). Tools are read-only unless a write scope is granted, members with a read-only role can choose only read scopes, API keys are limited to chosen environments, and there is no delete tool. No confirmation step for writes was found (15). Trace, log and span content is untrusted text, and no injection guidance was found in the MCP docs (2). The Activity Log records OAuth sessions with requested and granted scopes and one event per query run with its source, MCP included. The compliance page describes activity logging as for enterprise customers (11). SOC 2 Type II each year with penetration tests by an independent firm, HIPAA BAAs and an email disclosure route under a bug bounty page whose rewards are all at Honeycomb's discretion, so the bounty isn't counted. No security.txt, no stated acknowledgement time, and the scope names ui.honeycomb.io and api.honeycomb.io, not the MCP hosts. ISO 27001 is claimed only for the cloud hosts. One certification and a disclosure route, set between mailersend's 8 (disclosure route, discretionary bounty, no certification) and daytona's 15 (disclosure route, SOC 2 and ISO 27001, discretionary reward not counted) (12).\n- New Relic MCP Server (76): OAuth 2.0 with the scopes `observability:read`, `offline` and `offline_access`, access tokens that expire after 5 minutes, refresh for 24 hours, revocation by deleting the user or turning off the organisation toggle, and the token acting as the user with no more than their roles; the alternative is a user API key in a header (28). A read scope, organisation toggles that default the API key, preview, Atlassian and AWS routes to off, an organisation-scoped role requirement, and no write tools in the list (16). Logs, error messages and trace attributes are untrusted text and no injection guidance was found (2). NrAuditEvent records sign-ins with IP addresses and NrdbQuery labels queries sent through the server, both queryable by the operator (13). security.txt pointing to a HackerOne programme, a published security policy citing SOC 2 Type 2 and ISO 27001 audits of its facilities and its own penetration testing, and an ISO 42001 certification announced on 29 September 2026; the trust centre itself renders through a script and was unread (17). 28 + 16 + 2 + 13 + 17 = 76.\n\n### Schema \u0026 documentation: Honeycomb MCP +6\n\n- Honeycomb MCP (57): The server is closed source and `tools/list` needs sign-in, so we could not read the input schemas. Partial credit for a contract that exists but is not public (12). docs.honeycomb.io publishes llms.txt and a Markdown twin of every page (10). The tools reference says what each of the 36 tools returns and when an agent reaches for it, for example calling `get_workspace_context` first and `list_memories` before `read_memory`. These are docs, not the definitions the model sees (14). Parameters are mentioned only in passing, with no types, enums or required fields published (4). A use-cases page and a troubleshooting page document the Unauthorized, rate-limit, permission and not-found errors with causes and fixes (9). The registry entry carries version 1.0.0 and Honeycomb links a UI changelog and an API changelog, but changelog.honeycomb.io still disallows unnamed crawlers, so we could not confirm MCP entries (8).\n- New Relic MCP Server (51): Tool input schemas are only visible through an authenticated session; the tool reference names 40 tools with one-line descriptions, so partial credit for a contract that exists but isn't public (10). docs.newrelic.com publishes llms.txt, updated 3 November 2026, and a Markdown twin of every page (10). Each tool's line says what it does, tags say which group it belongs to, and the reference says which tools need OAuth and which are preview; nothing says when not to use a tool (12). No parameters, types or enums are published; the setup examples show only the `include-tags` header (4). Example natural-language queries per group, a troubleshooting page for OAuth, API key, permission and connection errors, and no error code list (7). A dated what's-new feed and a 0.1.0 registry version, but no MCP changelog or server version in the docs (8). 10 + 10 + 12 + 4 + 7 + 8 = 51.\n\n### Agent ergonomics: New Relic MCP Server +6\n\n- Honeycomb MCP (53): 36 tools is above the 30 line (5). The `X-MCP-Toolset` header loads `query` or `canvas` subsets, and the docs don't say how many tools each holds (add 8, so 13). Size controls are described per tool, such as up to 100 datasets or 100 columns a call, subtree zoom on `get_trace` and query limits. Pagination parameters aren't documented (10). Errors are documented with causes and fixes, and the rate-limit message names the retry time. No error code list was found (12). The reference labels each tool read or write and `update_slo` keeps omitted fields. No idempotency keys were found, and we couldn't confirm readOnlyHint or destructiveHint annotations (8). `get_workspace_context` takes no parameters, relative time is the default, and setup is one command or link in nine documented clients. SDKs don't apply (10).\n- New Relic MCP Server (59): 40 tools in the default list, above the 30-tool line (5), with the `include-tags` header to load only named groups (add 10, so 15). Time windows on most tools, `convert_time_period_to_epoch_ms`, filters on incidents, errors and logs, and NRQL's own LIMIT and FACET; no documented page size (14). The troubleshooting page covers setup errors with fixes, and the docs say nothing about tool-level error bodies (8). Every tool is a read (query, list, fetch, analyse) and the gap tool reads files the client sends, so a retry is safe; no readOnlyHint or destructiveHint was confirmed because tools/list wasn't read (12). Natural-language prompts need no parameters from the user, and the API behind the server has official NerdGraph clients; nothing agent-specific beyond skills (10). 15 + 14 + 8 + 12 + 10 = 59.\n\n## Score by category\n\n| Category | Weight | Honeycomb MCP | New Relic MCP Server | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 70 | 60 | Honeycomb MCP +10 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 57 | 51 | Honeycomb MCP +6 |\n| Agent ergonomics | 13% (16.2 this run) | 53 | 59 | New Relic MCP Server +6 |\n| Security \u0026 auth | 14% (17.5 this run) | 68 | 76 | New Relic MCP Server +8 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 30 | 30 | even |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 81 | 80 | Honeycomb MCP +1 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 70 | 81 | New Relic MCP Server +11 |\n| Negative events | ≤15 | 0 | 0 | |\n| **Total** | | **60.7 · C** | **61 · C** | |\n\n## Facts side by side\n\n| Fact | Honeycomb MCP | New Relic MCP Server |\n| --- | --- | --- |\n| Kind | MCP server | MCP server |\n| Vendor | Honeycomb | New Relic, Inc. |\n| Hosted endpoint | `https://mcp.honeycomb.io/mcp` | `https://mcp.newrelic.com/mcp/` |\n| Transports | Streamable HTTP | Streamable HTTP |\n| Auth | OAuth or key | OAuth |\n| Pricing | Your plan | Freemium |\n| x402 | no | no |\n| Licence | Proprietary hosted service under Honeycomb's Terms of Service and Honeycomb Intelligence Supplemental Terms. The companion plugin in honeycombio/agent-skill is MIT | none |\n| Tools exposed | 36 | 40 |\n| Read-only variant documented | no | yes |\n| llms.txt | yes | yes |\n| MCP registry | `io.honeycomb/mcp` | `com.newrelic/mcp-server` |\n| Last release | 2026-10-09 | 2026-10-28 |\n| Terms last updated | 2025-05-22 | 2022-11-08 |\n| Privacy policy last updated | no date given | 2026-09-14 |\n| Customer content may train models | not found in the text | not found in the text |\n| Terms restrict automated access | not found in the text | not found in the text |\n| Terms restrict benchmarking | not found in the text | yes |\n| Terms or service can change without notice | yes | not found in the text |\n| Arbitration or class-action waiver | not found in the text | not found in the text |\n| Popularity | 27 stars | none |\n\n## How the scores have moved\n\n### Honeycomb MCP\n\n- +2.2 on 4 November 2026: product, correction\n- +0.2 on 27 October 2026: checklist\n\n### New Relic MCP Server\n\nNo score change since it was listed.\n\n## Verdicts\n\n**Honeycomb MCP.** OAuth now lets an owner grant read or write per resource, so an agent can be held to queries and reads across the 36 tools, and per-tool rate limits are published. Tool input schemas are visible only after sign-in, no injection guidance was found for telemetry content, and the product changelogs disallow unnamed crawlers.\n\n**New Relic MCP Server.** OAuth with a read scope, 5-minute access tokens, role-bound calls and a queryable audit trail make this a safe read surface over an account's data, and the free tier includes it. 40 tools need tag filtering, the schemas aren't public, there is no per-call price, and no dated GA notice was found.\n\n## Before you call either\n\n### Honeycomb MCP\n\n1. Use https://mcp.honeycomb.io/mcp for US teams and https://mcp.eu1.honeycomb.io/mcp for EU teams. A team's data lives in one region only\n2. If no tools appear after sign-in, ask a Team Owner to enable Honeycomb Intelligence in team settings\n3. Ask the owner to choose Customize on the consent screen and grant only the resources the task needs. The agent sees only the tools its scopes allow\n4. Send `X-MCP-Toolset: query` or `canvas` to load a subset of the 36 tools. Call `get_workspace_context` first, then `find_columns`, before `run_query`\n5. For unattended runs send `Authorization: Bearer \u003cKEY_ID\u003e:\u003cSECRET_KEY\u003e` from a Management key with only Model Context Protocol and Environments scopes\n\n### New Relic MCP Server\n\n1. Connect with OAuth; an `api-key` header reaches only the core tools and the docs say the API key route works only where stated\n2. Send `include-tags` with the groups the task needs (for example `discovery,data-access`) to keep the tool list under 30\n3. Call `list_available_new_relic_accounts` first and pass the account id to every query; `convert_time_period_to_epoch_ms` turns phrases such as last 30 minutes into the epoch bounds the other tools take\n4. Stay under 2,000 tool calls an hour per user and 25 concurrent NerdGraph requests; a 429 clears as in-flight requests finish\n5. Ask the organisation administrator to turn on the Preview Tools, API Keys, Atlassian or AWS toggles in feature control manager before relying on those tools; they are off by default\n\n## Questions\n\n### Which is better for AI agents, Honeycomb MCP or New Relic MCP Server?\n\nNew Relic MCP Server and Honeycomb MCP score within a point of each other on agent readiness, 61 (C) and 60.7 (C). Honeycomb MCP leads on reliability and schema \u0026 documentation.\n\n### Do Honeycomb MCP and New Relic MCP Server need an API key?\n\nHoneycomb MCP takes an API key or an OAuth sign-in. New Relic MCP Server uses an OAuth sign-in.\n\n### Can an agent call Honeycomb MCP and New Relic MCP Server without installing anything?\n\nYes. Honeycomb MCP has a hosted endpoint at https://mcp.honeycomb.io/mcp and New Relic MCP Server at https://mcp.newrelic.com/mcp/.\n\n\n## For agents\n\n- This comparison as JSON: https://www.anchorterminal.com/compare/honeycomb-mcp-vs-new-relic.json, and with the fewest tokens: https://www.anchorterminal.com/compare/honeycomb-mcp-vs-new-relic.min.md\n- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {\"a\": \"honeycomb-mcp\", \"b\": \"new-relic\"}`. From a terminal: `anchor compare honeycomb-mcp new-relic`\n- Each listing in full: https://www.anchorterminal.com/api/v1/tools/honeycomb-mcp.json and https://www.anchorterminal.com/api/v1/tools/new-relic.json\n\n## Other comparisons with Honeycomb MCP or New Relic MCP Server\n\n- [Causely MCP server vs New Relic MCP Server](https://www.anchorterminal.com/compare/causely-vs-new-relic.md)\n- [incident.io API vs New Relic MCP Server](https://www.anchorterminal.com/compare/incident-io-vs-new-relic.md)\n- [New Relic MCP Server vs PagerDuty MCP Server](https://www.anchorterminal.com/compare/new-relic-vs-pagerduty-mcp.md)\n- [New Relic MCP Server vs Rootly MCP Server](https://www.anchorterminal.com/compare/new-relic-vs-rootly-mcp.md)\n- [Axiom MCP Server vs Honeycomb MCP](https://www.anchorterminal.com/compare/axiom-vs-honeycomb-mcp.md)\n- [Axiom MCP Server vs New Relic MCP Server](https://www.anchorterminal.com/compare/axiom-vs-new-relic.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- [Causely MCP server vs Honeycomb MCP](https://www.anchorterminal.com/compare/causely-vs-honeycomb-mcp.md)\n- [Datadog MCP Server vs Honeycomb MCP](https://www.anchorterminal.com/compare/datadog-mcp-vs-honeycomb-mcp.md)\n- [Datadog MCP Server vs New Relic MCP Server](https://www.anchorterminal.com/compare/datadog-mcp-vs-new-relic.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- [New Relic MCP Server vs Sentry MCP](https://www.anchorterminal.com/compare/new-relic-vs-sentry-mcp.md)\n- [Honeycomb MCP vs Sentry MCP](https://www.anchorterminal.com/compare/honeycomb-mcp-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": "Honeycomb MCP vs New Relic MCP Server", "url": "" } ], "description": "New Relic and Honeycomb MCP score within a point of each other for observability traces, 61 and 60.7 out of 100. Prices, MCP, x402, uptime and agent notes side by side.", "facts": [ "Honeycomb MCP C 60.7", "New Relic MCP Server C 61", "scores" ], "h1": "Honeycomb MCP vs New Relic MCP Server", "image": "https://www.anchorterminal.com/assets/og/default.png", "path": "/compare/honeycomb-mcp-vs-new-relic", "published": "2026-10-01", "section": "tools", "title": "Honeycomb MCP vs New Relic for AI agents in 2026: scores and prices", "toc": null, "updated": "2026-11-04", "url": "https://www.anchorterminal.com/compare/honeycomb-mcp-vs-new-relic" }, "tokens": { "markdown": 5250, "slim": 3430 }, "version": 1 }