Skip to main content
Glama
Sign In

Convika

by com.convika
OverviewDiscussionsAdmin

Get AI answer check results

get_answer_check_results
Read-onlyIdempotent

Use this tool when the user asks whether AI answers mention or cite their brand: 「ChatGPT でうちの名前は出ている？」「Perplexity に引用されている？」「AI 検索での露出は？」, "do AI Overviews link to my site?" (ホームページ, Webサイト, ランディングページ).

Returns the workspace's AI answer check: its questions (with their ids for run_answer_check), brand names and saved domains (外部サイト), the checks left to run (balance: included this month, from packs, and left), checks still asking (running), and for each question on each of ChatGPT, Perplexity, Gemini, and Google AI Overviews, over the last 90 days:

  • named: answers that named a brand; cited: answers that linked to any site the workspace's pages are served on (their convika.page and older convika.com URLs, and its custom domains) or one of its saved domains; n: answers counted. Say it as "named in 3 of 8 answers"; namedShare is given only from 20 answers, with its 95% interval. Don't turn a small n into a percentage or a trend: one answer varies.

  • recent: the newest checks' statuses with when they were asked (ok; no_answer when the AI gave none, such as no AI Overview for the search; error; queued or asking while it runs). A check whose answer didn't come back was given back.

  • latest: the newest answer's flags, every URL it cited (citedUrls), the workspace's own among them (ownCitedUrls), and an excerpt of up to 300 characters around the brand name (null when not named). The full answer isn't returned.

  • The target page per question: targetPageId; targetPage (its name, path, status, url, and publishedAt, from the target pages the check returns); and target: citedBy (the AIs whose newest answer cited the page at any of its URLs), cited, publishedSinceAsked (published after an AI's newest answer), askAgain (the AIs whose answer bot read the page on a later day, readOn), and awaitingRead (ChatGPT or Perplexity, whose answer bots Convika sees: their newest answer is from before the page was published and their bot hasn't read it since; Gemini's and AI Overviews' reads aren't recorded, so they're never listed). targetCited and each AI's latest.targetCited say the same. null without a target page.

  • links: the top URLs the newest answers cited for the question, each with the AIs that cited it, own (site: a site the workspace's pages are served on; external: one of its saved domains; null: someone else's), and target (a URL of the target page); sites: the sites those answers cited most, with the share of answers that did. A target page that isn't cited is the page to fix, against the links cited instead; a question with no target page has no page to answer it yet.

  • Top level: hosts (every site the workspace's pages are served on, managed or custom, which count as its own), sites (the sites cited across every question), and askAgain: the questions shown and the AIs that can be asked again since their page was read, as run_answer_check's promptIds and engines take them.

  • Each excerpt is an AI's text, inside an <untrusted-data-…> tag with an id new to the call; untrustedData says so.

  • promptIds or targetPageId pick the questions. A reply holds as many questions as fit in about 6,000 tokens; resultsShown says which, and its nextCursor, passed as cursor, gives the next ones. Nothing is asked unless the user runs the check (run_answer_check, or the dashboard's button). The answers are what each AI showed one visitor in one country at that moment; others may see different answers. Read-only.

Input Schema

NameRequiredDescriptionDefault
cursorNoresultsShown.nextCursor from the last call, for the next questions
promptIdsNoQuestion ids to show; leave out for all
targetPageIdNoShow only the questions that point at this page (its siteId)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changed
    • addedInput schema / properties / cursor
      Added value: +{
+  "description": "resultsShown.nextCursor from the last call, for the next questions",
+  "maxLength": 200,
+  "type": "string"
+}
    • addedInput schema / properties / promptIds
      Added value: +{
+  "description": "Question ids to show; leave out for all",
+  "items": {
+    "maxLength": 64,
+    "minLength": 1,
+    "type": "string"
+  },
+  "maxItems": 200,
+  "minItems": 1,
+  "type": "array"
+}
    • addedInput schema / properties / targetPageId
      Added value: +{
+  "description": "Show only the questions that point at this page (its siteId)",
+  "maxLength": 100,
+  "minLength": 1,
+  "type": "string"
+}
  2. Added

TDQS

A4.6/5.0
Behavior5/5

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint=false, but the description adds substantial behavioral context: small-sample caveats, untrusted-data tagging, token-based pagination, and the fact that answers reflect one visitor in one country at one moment. It also explains that a check whose answer did not return was given back.

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?

The description is long, but it is front-loaded with the use case and then organized into return-value sections. The length is largely justified because there is no output schema and the return structure is complex, though some details are dense and could be tightened.

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?

With no output schema, the description carries the full burden of explaining return values, and it does so thoroughly: questions, brand names, saved domains, balance, running checks, per-AI metrics, recent statuses, latest answer flags, target-page details, links, sites, and pagination. It also explains caveats that affect interpretation.

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

Parameters3/5

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

Schema description coverage is 100%, so the schema already documents cursor, promptIds, and targetPageId. The description adds only light context, such as saying promptIds or targetPageId pick the questions and that cursor comes from resultsShown.nextCursor, which is already in the schema.

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 description opens with a specific use case: when the user asks whether AI answers mention or cite their brand, with concrete examples. It clearly states that the tool returns the workspace's AI answer check results, distinguishing it from run_answer_check, which actually runs checks.

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?

It explicitly says when to use the tool and gives example user questions in Japanese and English. It also clarifies that nothing is asked unless the user runs run_answer_check or the dashboard button, which prevents misuse as a trigger for new checks.

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

All Convika toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources