Skip to main content
Glama

Get AI search visibility numbers for a date range

visibility_get_summary
Read-onlyIdempotent

Returns the numbers behind the Visibility dashboard in Hi, Moose for one date range, with the same filters the dashboard offers: topics, engines, prompts, branded or non-branded prompts, pages cited, sentiment, narrative, feature parity, plus mentioned, cited and competitor. Pick sections to keep the result small: totals (always), engines, topics, prompts, trend (per day), competitors (share of voice, per engine and topic, pages competitors get cited from, prompts competitors win), citations (pages and domains the engines cite with counts per engine, own and competitor domains flagged, paged with citationsLimit and citationsOffset), mentions (every brand named, with rank and sentiment). Pass compare true for the equal window before, or compareFrom and compareTo, to get the same totals for a second window. Use this for reports, dashboards and "how are we doing" questions; use visibility_list_results for the individual answers behind a number and visibility_check_topic for a one-topic question in plain words. Counts cover the tracked prompts on the tracked engines only, sampled on run days, never the whole of AI search: report the window and sample size and never claim causation. Reads stored results only. Takes a few seconds.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "project": "<project>"
    },
    "name": "visibility_get_summary"
  }
}

Input Schema

NameRequiredDescriptionDefault
toNoLast day of the window as YYYY-MM-DD. Default today.
fromNoFirst day of the window as YYYY-MM-DD. Default: lookbackDays before to.
citedNotrue: only answers that cite a page of the brand. false: only answers that do not.
pagesNoOnly prompts whose answers cited one of these pages, then every engine for those prompts. Takes page URLs or bare domains such as "example.com", the brand's own or any other site, as returned in the citations section.
topicsNoOnly these topics, by the exact names from visibility_list_tracked. "Uncategorized" selects prompts with no topic.
compareNotrue: also return totals for the window of the same length immediately before from.
enginesNoOnly these AI engines, by name from visibility_list_tracked, such as "ChatGPT" or "Perplexity". A plain name also matches its local-fetch variant.
projectYesThe Hi, Moose project, by its name as shown by list_projects. The projectId is also accepted.
promptsNoOnly these prompts, by exact text or promptId from visibility_list_tracked.
sectionsNoWhich sections to return. Default totals, engines and topics.
compareToNoLast day of a custom comparison window, YYYY-MM-DD. Needs compareFrom.
mentionedNotrue: only answers that mention the brand. false: only answers that do not.
narrativeNoOnly answers with this narrative verdict against the Brand Truth Profile: aligned, partial or drifted.
sentimentNoOnly answers where the brand was mentioned with this sentiment.
competitorNoOnly answers that mention this competitor, by name from visibility_list_tracked.
promptTypeNoOnly branded prompts (the prompt names the brand, an alias or an owned domain) or only non-branded prompts.
compareFromNoFirst day of a custom comparison window, YYYY-MM-DD. Needs compareTo.
citedDomainsNocitations section: return exact counts for only these domains (and their subdomains) and the pages on them, however rarely they were cited. Unlike pages, this does not change which answers are counted.
lookbackDaysNoWindow length in days when from is not given. Default 30.
minCitationsNocitations section: only domains and pages cited at least this many times in the window.
citationsListNocitations section: which list to return. Default both. Ask for domains or pages alone to fit more rows in one result.
featureParityNoOnly answers with this feature parity verdict: accurate, needs-review, inaccurate or not-applicable.
citationsLimitNocitations section: how many rows of each list to return per call, most cited first. Default 50. Each row carries per-engine counts, so keep the rows in one result near 150 or fewer (for example 150 of one list, or 75 of both) and page with citationsOffset for more.
citationsOffsetNocitations section: skip this many domains and pages, to page through the full lists. Pass the nextOffset of the previous call. Default 0.
mergeEngineVariantsNocitations section: count the local-fetch variant of an engine under the engine's plain name in byEngine, so "ChatGPT (local fetch)" counts as "ChatGPT". Default true.

Output Schema

NameRequiredDescriptionDefault
errorNo
jobIdNo
rangeNo
trendNo
statusNodone, failed, pending or running.
totalsNoCounts over the answers in scope. Rates are fractions of answers (0 to 1) and null when there are none. unavailable answers are excluded from the rates.
byTopicNo
filtersNoThe filters that were applied, after validation.
byEngineNo
byPromptNo
headlineNoOne sentence with the window, sample size and the main numbers.
mentionsNo
progressNo
runCountNo
sectionsNo
citationsNo
createdAtNo
comparisonNo
dataNoticeNo
competitorsNo
projectNameNo
promptCountNo
byTopicEngineNo
schemaVersionNoVersion of this result shape. Fields are only added within a version.
trendByEngineNo
byPromptTruncatedNotrue when more than 300 prompts matched; filter by topic to see the rest.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changed
    • changedInput schema / properties / citationsLimit / description
      Previous value: -"citations section: how many domains and how many pages to return per call, most cited first. Default 200."New value: +"citations section: how many rows of each list to return per call, most cited first. Default 50. Each row carries per-engine counts, so keep the rows in one result near 150 or fewer (for example 150 of one list, or 75 of both) and page with citationsOffset for more."
    • changedInput schema / properties / citationsLimit / maximum
      Previous value: -500New value: +150
    • addedInput schema / properties / citationsList
      Added value: +{
      +  "description": "citations section: which list to return. Default both. Ask for domains or pages alone to fit more rows in one result.",
      +  "enum": [
      +    "both",
      +    "domains",
      +    "pages"
      +  ],
      +  "type": "string"
      +}
    • addedOutput schema / properties / citations / properties / list
      Added value: +{
      +  "description": "Which list this result holds. The other list is empty when one was asked for alone.",
      +  "enum": [
      +    "both",
      +    "domains",
      +    "pages"
      +  ],
      +  "type": "string"
      +}
  2. Changed20 schema fields changed
    • addedInput schema / properties / citationsLimit
      Added value: +{
      +  "description": "citations section: how many domains and how many pages to return per call, most cited first. Default 200.",
      +  "maximum": 500,
      +  "minimum": 1,
      +  "type": "integer"
      +}
    • addedInput schema / properties / citationsOffset
      Added value: +{
      +  "description": "citations section: skip this many domains and pages, to page through the full lists. Pass the nextOffset of the previous call. Default 0.",
      +  "minimum": 0,
      +  "type": "integer"
      +}
    • addedInput schema / properties / citedDomains
      Added value: +{
      +  "description": "citations section: return exact counts for only these domains (and their subdomains) and the pages on them, however rarely they were cited. Unlike pages, this does not change which answers are counted.",
      +  "items": {
      +    "maxLength": 300,
      +    "type": "string"
      +  },
      +  "maxItems": 200,
      +  "type": "array"
      +}
    • addedInput schema / properties / mergeEngineVariants
      Added value: +{
      +  "description": "citations section: count the local-fetch variant of an engine under the engine's plain name in byEngine, so \"ChatGPT (local fetch)\" counts as \"ChatGPT\". Default true.",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / minCitations
      Added value: +{
      +  "description": "citations section: only domains and pages cited at least this many times in the window.",
      +  "minimum": 1,
      +  "type": "integer"
      +}
    • changedInput schema / properties / pages / description
      Previous value: -"Only prompts whose answers cited one of these page URLs, then every engine for those prompts, which is how the Page filter works in Hi, Moose."New value: +"Only prompts whose answers cited one of these pages, then every engine for those prompts. Takes page URLs or bare domains such as \"example.com\", the brand's own or any other site, as returned in the citations section."
    • addedOutput schema / properties / citations / properties / citedDomains
      Added value: +{
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedOutput schema / properties / citations / properties / domains / description
      Added value: +"Cited domains, most cited first, one page of the list (see limit, offset, domainsTotal). own marks the brand's domains; competitor holds the tracked competitor's name when the domain is theirs, else null. byEngine holds the counts for each AI engine. Tracker and AI engine app domains are left out."
    • addedOutput schema / properties / citations / properties / domains / items / properties / byEngine
      Added value: +{
      +  "additionalProperties": {
      +    "properties": {
      +      "answers": {
      +        "type": "integer"
      +      },
      +      "citations": {
      +        "type": "integer"
      +      },
      +      "prompts": {
      +        "type": "integer"
      +      }
      +    },
      +    "type": "object"
      +  },
      +  "description": "Counts per AI engine, keyed by engine name.",
      +  "type": "object"
      +}
    • addedOutput schema / properties / citations / properties / domains / items / properties / competitor
      Added value: +{
      +  "type": [
      +    "string",
      +    "null"
      +  ]
      +}
    • addedOutput schema / properties / citations / properties / domainsTotal
      Added value: +{
      +  "description": "How many domains match in all, across every page of the list.",
      +  "type": "integer"
      +}
    • addedOutput schema / properties / citations / properties / enginesMerged
      Added value: +{
      +  "description": "true when local-fetch variants are counted under the plain engine name in byEngine.",
      +  "type": "boolean"
      +}
    • addedOutput schema / properties / citations / properties / limit
      Added value: +{
      +  "type": "integer"
      +}
    • addedOutput schema / properties / citations / properties / minCitations
      Added value: +{
      +  "type": "integer"
      +}
    • addedOutput schema / properties / citations / properties / nextOffset
      Added value: +{
      +  "description": "Pass as citationsOffset to get the next domains and pages. null when both lists are complete.",
      +  "type": [
      +    "integer",
      +    "null"
      +  ]
      +}
    • addedOutput schema / properties / citations / properties / offset
      Added value: +{
      +  "type": "integer"
      +}
    • addedOutput schema / properties / citations / properties / pages / description
      Added value: +"Cited pages, most cited first, one page of the list (see limit, offset, pagesTotal). byEngine holds the counts for each AI engine."
    • addedOutput schema / properties / citations / properties / pages / items / properties / byEngine
      Added value: +{
      +  "additionalProperties": {
      +    "properties": {
      +      "answers": {
      +        "type": "integer"
      +      },
      +      "citations": {
      +        "type": "integer"
      +      },
      +      "prompts": {
      +        "type": "integer"
      +      }
      +    },
      +    "type": "object"
      +  },
      +  "description": "Counts per AI engine, keyed by engine name.",
      +  "type": "object"
      +}
    • addedOutput schema / properties / citations / properties / pages / items / properties / competitor
      Added value: +{
      +  "type": [
      +    "string",
      +    "null"
      +  ]
      +}
    • addedOutput schema / properties / citations / properties / pagesTotal
      Added value: +{
      +  "description": "How many pages match in all, across every page of the list.",
      +  "type": "integer"
      +}
  3. First observed

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare read-only, idempotent and non-destructive, but the description adds material context beyond them: it reads stored results only, takes a few seconds, and the counts are sampled on run days over tracked prompts and engines only — with an explicit instruction to report the window and sample size and never claim causation. That sampling limitation is a genuine behavioral trait an agent could not infer from the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose is front-loaded and nearly every sentence carries load, but the body is a single dense paragraph with no breaks, mixing section enumeration, comparison options, sibling routing and sampling caveats. The section list partially duplicates the enum in the schema and could have been structured more scannably for a 25-parameter tool.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a high-parameter-count aggregation tool this is complete: purpose, section selection, comparison windows, pagination, routing to siblings and the sampling/claim caveats are all present, and an output schema exists so return shapes need no explanation in prose.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description earns more: it explains what each selectable section actually contains (competitors = share of voice plus pages cited from plus prompts won; mentions = every brand with rank and sentiment), how compare/compareFrom/compareTo produce a second window, and how citationsLimit/citationsOffset page through results with a practical row-count heuristic. This meaningfully supplements the schema rather than restating it.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening sentence states a specific verb and resource ('Returns the numbers behind the Visibility dashboard ... for one date range') and immediately scopes it with the same filters the dashboard exposes. It explicitly separates this tool from visibility_list_results (individual answers behind a number) and visibility_check_topic (one-topic plain-language question), so an agent can route without opening another schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description names the exact use cases ('reports, dashboards and "how are we doing" questions') and then names two alternatives with the condition that selects each. It also gives a narrowing instruction ('Pick sections to keep the result small') and a reporting caveat, leaving little to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources