Bitmovin Observability MCP server
com.bitmovin.mcp.analytics/analytics-mcpanalytics.mcp.bitmovin.comstreamable-httpv0.0.1Scanned
Query and analyse your data from Bitmovin Observability
11 tools: 11 only read.
- Reads private data · 0
- Reads text others write · 11
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
Decide which of this server's tools your agent can use.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Bitmovin Observability MCP server can do
The Bitmovin Observability MCP server is published by analytics.mcp.bitmovin.com, a domain the MCP Registry verified. It has 11 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
11 tools bring in text other people can write (query, queryGroupBy, queryTotal and 8 more).
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 12, 2026, version 0.0.1.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
11 tools
Only reads 11Show 11 read-only tools
analyzeImpressionDeep-dive analysis of a SINGLE playback session identified by a specific impressionId: static properties, aggregated metrics, state-transition timeline, and an AI-powered interpret…Reads text others writeNo findings
Why it is in this group
- The name "analyzeImpression" says "analyze": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Deep-dive analysis of a SINGLE playback session identified by a specific impressionId: static properties, aggregated metrics, state-transition timeline, and an AI-powered interpretation of session quality. Use for detailed single-session troubleshooting. REQUIRES an impressionId. Do NOT use this for license-wide metrics, totals, trends, or any time-range ('last week', start/end) analysis — this tool does not accept start/end. For metrics over a time period use the 'query' tool; for breakdowns by category use 'queryGroupBy'.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
fetchImpressionsList the impression ids (individual playback session ids) for a license over a time range, optionally narrowed by filters.Reads text others writeNo findings
Why it is in this group
- The name "fetchImpressions" says "fetch": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it List the impression ids (individual playback session ids) for a license over a time range, optionally narrowed by filters. Returns the ids themselves — NOT counts, totals, or trends. Use this to enumerate sessions before drilling into specific ones with `getImpressionOverview` or `analyzeImpression`. Typical uses: 'give me impression ids from yesterday', 'list sessions on Safari that errored last week', 'sample sessions from country US'. Every matching session is eligible (including failed/setup sessions), not only successful plays. Results are capped (default 100, max 200); if truncated, narrow the timeframe/filters or raise `limit`. A licenseKey is required, and `peekAllLicenses` lists the available ones; `searchFilters` resolves the filter attributes and operators used by the optional `filters` input. For metric counts/trends over a period use `query` / `queryTotal`; for category breakdowns use `queryGroupBy`.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getAvailableFiltersList the full catalog of filter attributes with their supported operators.Reads text others writeNo findings
Why it is in this group
- The name "getAvailableFilters" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it List the full catalog of filter attributes with their supported operators. Prefer searchFilters to resolve a single attribute; use this only when you need every filter.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getAvailableMetricsList the full catalog of metrics with their supported aggregation methods.Reads text others writeNo findings
Why it is in this group
- The name "getAvailableMetrics" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it List the full catalog of metrics with their supported aggregation methods. Prefer searchMetrics to resolve a single keyword; use this only when you need every metric.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
getImpressionOverviewThe facts of a SINGLE playback session identified by an impressionId, together with the findings evaluated over them.Reads text others writeNo findings
Why it is in this group
- The name "getImpressionOverview" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it The facts of a SINGLE playback session identified by an impressionId, together with the findings evaluated over them. Facts cover the session context (device, location, player, stream, CDN), its QoE numbers (startup time, rebuffering, bitrates, quality changes, dropped frames), time spent per state, errors placed in the phase they arrived in, ad data, and the collector limits that make any of these incomplete. Findings rate startup, quality switching, rebuffering, errors by phase and start failures, flag the collector limits, and state what two co-occurring findings mean together. Each carries a severity, evidence paths pointing back at the facts it was derived from, and an explanation covering the mechanism behind it, the causes one session cannot separate, and what commonly addresses it. They are computed from fixed thresholds, so the same session always yields the same findings; `analyzeImpression` adds an AI-written interpretation of the same session. An impressionId is required, and `fetchImpressions` lists the ids for a license and timeframe. The tool takes no start/end and covers one session: metrics over a time period come from `query` / `queryTotal`, breakdowns by category from `queryGroupBy`.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
peekAllLicensesView a summary of all available licenses with their recent play counts and percentage distribution.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "view": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it View a summary of all available licenses with their recent play counts and percentage distribution. Use this to understand what licenses exist and their relative usage. When a user asks about licenses by ranking (e.g., 'biggest', 'most active', 'first'), use this to identify which license matches that criteria before querying.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
queryQuery a metric over a time range (start/end) for a license as a TIME SERIES — always renders a chart.Reads text others writeNo findings
Why it is in this group
- The name "query" says "query": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Query a metric over a time range (start/end) for a license as a TIME SERIES — always renders a chart. Use this for trends and license-wide analytics over a period (e.g. 'last week', 'last 30 days', 'yesterday'). Bucket size is chosen automatically from start/end; there is no interval field to set. For a single aggregated number (one total/average over the whole period, NO chart), use the separate `queryTotal` tool instead. Queries resolve against API metric keywords, which can differ from a metric's common name — 'plays', for example, is 'impression_id'. The 'searchMetrics' tool resolves a keyword cheaply; 'getAvailableMetrics' returns the full catalog. Every query needs a licenseKey; `peekAllLicenses` lists the ones available. AD ANALYTICS: ad metrics use 'ad_'-prefixed keywords and are queried against a separate ads dataset. The ad completion funnel is: 'ad_quartile_1' (first quartile / 25%) -> 'ad_midpoint' (50%) -> 'ad_quartile_3' (third quartile / 75%) -> 'ad_completions'. Other ad metrics: 'ad_impressions', 'ad_starts', 'ad_clicks', 'ad_skips', 'ad_startup_time', 'ad_error_percentage', 'ad_error_sessions', 'ad_time_played', 'ad_play_percentage', 'ad_unique_users'. Ad data is present on some licenses and not others; `peekAllLicenses` output shows which. A query covers a single license, so an ad question resolves against one license at a time. With an ad metric, filters and groupBy take the AD-specific attributes from 'searchFilters' / 'getAvailableFilters' (e.g. 'AD_SYSTEM', 'AD_POSITION', 'ADVERTISER_NAME'); playback attributes and ad attributes are not interchangeable. OPTIONAL 3) Use the 'searchFilters' tool to find filter attributes/operators (or 'getAvailableFilters' for the full list).
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
queryGroupByquery metric grouped by categories.Reads text others writeNo findings
Why it is in this group
- The name "queryGroupBy" says "query": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it query metric grouped by categories. Queries resolve against API metric keywords, which can differ from a metric's common name — 'plays', for example, is 'impression_id'. The 'searchMetrics' tool resolves a keyword cheaply; 'getAvailableMetrics' returns the full catalog. Every query needs a licenseKey; `peekAllLicenses` lists the ones available. AD ANALYTICS: ad metrics use 'ad_'-prefixed keywords and are queried against a separate ads dataset. The ad completion funnel is: 'ad_quartile_1' (first quartile / 25%) -> 'ad_midpoint' (50%) -> 'ad_quartile_3' (third quartile / 75%) -> 'ad_completions'. Other ad metrics: 'ad_impressions', 'ad_starts', 'ad_clicks', 'ad_skips', 'ad_startup_time', 'ad_error_percentage', 'ad_error_sessions', 'ad_time_played', 'ad_play_percentage', 'ad_unique_users'. Ad data is present on some licenses and not others; `peekAllLicenses` output shows which. A query covers a single license, so an ad question resolves against one license at a time. With an ad metric, filters and groupBy take the AD-specific attributes from 'searchFilters' / 'getAvailableFilters' (e.g. 'AD_SYSTEM', 'AD_POSITION', 'ADVERTISER_NAME'); playback attributes and ad attributes are not interchangeable. The groupBy attribute is an exact API attribute name, and common guesses are often wrong: 'OPERATING_SYSTEM' is invalid where 'OPERATINGSYSTEM' is correct, and an unrecognised attribute fails with a validation error. The 'searchFilters' tool resolves an attribute name; 'getAvailableFilters' returns the full catalog. Results grouped by ERROR_CODE carry occurrence counts but not error semantics, which are Bitmovin-specific and documented in the live Bitmovin documentation served by the docs MCP tool (typically `general_docs_ask_bitmovin_docs`). That tool resolves one numeric error code per call and returns nothing usable for a question naming several codes at once.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
queryTotalGet an aggregated value (one number) for a metric over a time range — no chart, no time bucketing.Reads text others writeNo findings
Why it is in this group
- The name "queryTotal" says "query": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Get an aggregated value (one number) for a metric over a time range — no chart, no time bucketing. Can fetch SEVERAL metrics in one call: pass an array to `metric` to get one value per metric (e.g. the ad completion funnel), instead of calling this tool repeatedly. Use ONLY when the user asks for totals/averages/median/p95 over a whole period (e.g. 'total plays last month', 'average startup time yesterday', 'p95 rebuffer last week', 'the ad funnel counts last week'). For trends or any 'how did X change over time' / 'show last week' question, use the `query` tool instead — it always renders a chart. Queries resolve against API metric keywords, which can differ from a metric's common name — 'plays', for example, is 'impression_id'. The 'searchMetrics' tool resolves a keyword cheaply; 'getAvailableMetrics' returns the full catalog. Every query needs a licenseKey; `peekAllLicenses` lists the ones available. AD ANALYTICS: ad metrics use 'ad_'-prefixed keywords and are queried against a separate ads dataset. The ad completion funnel is: 'ad_quartile_1' (first quartile / 25%) -> 'ad_midpoint' (50%) -> 'ad_quartile_3' (third quartile / 75%) -> 'ad_completions'. Other ad metrics: 'ad_impressions', 'ad_starts', 'ad_clicks', 'ad_skips', 'ad_startup_time', 'ad_error_percentage', 'ad_error_sessions', 'ad_time_played', 'ad_play_percentage', 'ad_unique_users'. Ad data is present on some licenses and not others; `peekAllLicenses` output shows which. A query covers a single license, so an ad question resolves against one license at a time. With an ad metric, filters and groupBy take the AD-specific attributes from 'searchFilters' / 'getAvailableFilters' (e.g. 'AD_SYSTEM', 'AD_POSITION', 'ADVERTISER_NAME'); playback attributes and ad attributes are not interchangeable.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
searchFiltersFind the exact filter/group-by attribute keyword for a topic or question (e.g.Reads text others writeNo findings
Why it is in this group
- The name "searchFilters" says "search": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Find the exact filter/group-by attribute keyword for a topic or question (e.g. 'browser', 'device type', 'content tier'). Preferred way to resolve an attribute; use 'licenseKey' to resolve custom namings of attributes; use getAvailableFilters only to list the full static catalog.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
searchMetricsFind the exact metric keyword for a topic or question (e.g.Reads text others writeNo findings
Why it is in this group
- The name "searchMetrics" says "search": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Find the exact metric keyword for a topic or question (e.g. 'buffering', 'how many errors'). Preferred way to resolve a metric keyword; use getAvailableMetrics only to list the full catalog.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write11
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:35 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 0
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- yes
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools1 changed.~1 changedView diff
- Changed:getImpressionOverviewOnly readsdescription
Install
Pinned to version 0.0.1. Values in angle brackets are yours to fill in.
claude mcp add --transport http analytics https://analytics.mcp.bitmovin.com --header 'x-api-key: <your-x-api-key>'claude mcp add --transport http analytics https://analytics.mcp.bitmovin.com --header 'x-api-key: <your-x-api-key>'Run in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__analytics__query",
"mcp__analytics__queryGroupBy",
"mcp__analytics__queryTotal",
"mcp__analytics__peekAllLicenses",
"mcp__analytics__getImpressionOverview",
"mcp__analytics__analyzeImpression",
"mcp__analytics__fetchImpressions",
"mcp__analytics__getAvailableMetrics",
"mcp__analytics__getAvailableFilters",
"mcp__analytics__searchMetrics",
"mcp__analytics__searchFilters"
],
"ask": [],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as analytics. Review it before you use it.
Capability tags behind 11 rules
- mcp__analytics__query untrusted-input
- mcp__analytics__queryGroupBy untrusted-input
- mcp__analytics__queryTotal untrusted-input
- mcp__analytics__peekAllLicenses untrusted-input
- mcp__analytics__getImpressionOverview untrusted-input
- mcp__analytics__analyzeImpression untrusted-input
- mcp__analytics__fetchImpressions untrusted-input
- mcp__analytics__getAvailableMetrics untrusted-input
- mcp__analytics__getAvailableFilters untrusted-input
- mcp__analytics__searchMetrics untrusted-input
- mcp__analytics__searchFilters untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/com.bitmovin.mcp.analytics/analytics-mcp/verify?hash=7c0f0c1ee7d2cbd1c095cabd446bc0edbf28ad3d279516272da783e7315a40df" | jq -e '.changed == false'Pinned hash 7c0f0c1ee7d2, from Nov 12, 2026, 18:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.bitmovin.mcp.analytics/analytics-mcp/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.bitmovin.mcp.analytics/analytics-mcp)Source: MCP Registry. Synced Oct 21, 2026, 17:51 UTC.
Bitmovin Observability MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Bitmovin Observability MCP server safe?
- No scan can promise that. What its definitions show: 11 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.
Can the Bitmovin Observability MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Bitmovin Observability MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the Bitmovin Observability MCP server have?
- 11 tools: 11 read. query, queryGroupBy, queryTotal, peekAllLicenses, getImpressionOverview, analyzeImpression, fetchImpressions, getAvailableMetrics and 3 more.
Does the Bitmovin Observability MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Bitmovin Observability MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Bitmovin Observability MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Oct 30, 2026. Every change is on this page and in its Atom feed.