Skip to main content
Glama

Refora

AI answers

get_ai_answers
Read-onlyIdempotent

The answers themselves: what ChatGPT, Perplexity, Gemini, Claude, Mistral and Google AI Overviews said to this workspace's prompts, each with the brands it mentions and the URLs it cites.

Use this tool when the user wants to read or quote what an AI said -- phrases like 'what does ChatGPT say about us', 'show me the answers to the prompt about X', 'how do AI models describe our competitor', 'show answers where we are not mentioned' -- or to explain a number another Refora tool returned. For numbers (visibility, share of voice, trends, sentiment, sources), use those tools: they count every answer, this one returns a page.

Filter by prompt (its text or a distinctive part of it), model, brand (brand_mentioned=true for answers that mention it, false for answers that don't; without brand_name it is the workspace's own brand) and window. Answers come newest first, 10 per page; when next_offset is set, call again with offset=next_offset. An answer longer than 2,000 characters is cut (truncated: true); to read it whole, call again with its prompt_text, model, its date as start_date and end_date, and full_text=true.

Example: 'What did Perplexity say when asked for the best CRM?'

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": {
      "workspace_id": "<workspace_id>"
    },
    "name": "get_ai_answers"
  }
}

Input Schema

NameRequiredDescriptionDefault
daysNoHow many days back to look, counted back from end_date (default today). Default 30, max 90. Ignored when start_date is given. Prompts are re-run on a schedule, so a window shorter than a week can be empty.
limitNoAnswers per page. Default 10, max 25. A page can hold fewer when its answers are long.
modelNoOptional: only this AI model's answers. openai = ChatGPT, anthropic = Claude, mistral = Le Chat, google-aio = Google AI Overviews. Omit for every model.
offsetNoHow many matching answers to skip. Pass the previous page's next_offset. Default 0.
end_dateNoOptional end of the window, as an ISO date (YYYY-MM-DD). Default today. Cannot be in the future.
full_textNotrue to return answers whole (up to 20,000 characters each) instead of cut at 2,000. Narrow the filters first: meant for reading one or a few answers. Default false.
brand_nameNoOptional: a brand to filter on. Alone, it keeps the answers that mention it; with brand_mentioned=false, the ones that don't. Omit it and set brand_mentioned to filter on the workspace's own brand.
start_dateNoOptional explicit start of the window, as an ISO date (YYYY-MM-DD). A full ISO datetime is accepted and its date is used. Overrides days.
prompt_textNoOptional: only answers to this prompt. Its text as another Refora tool returned it, or a distinctive part of it, matched case-insensitively. If several prompts match, the response lists them so the user can pick. Omit for every prompt.
workspace_idYesThe workspace's numeric id, as returned by list_workspaces.
brand_mentionedNoOptional: true for only answers that mention the brand, false for only answers that don't. The brand is brand_name, or the workspace's own brand. Omit for answers either way.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Beyond the readOnly/idempotent annotations, it discloses ordering (newest first), page size (10, configurable to 25), the continuation contract (call again with offset=next_offset), truncation at 2,000 characters with a truncated flag, and the exact recipe to retrieve full text. Those pagination and truncation behaviors are the main failure modes for this tool and are all stated.

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?

Front-loaded with what the tool returns, then usage guidance, then pagination mechanics, then an example — a sensible ordering. A few details (pagination, truncation) are duplicated between description and schema field docs, and the closing example is a nice-to-have rather than essential.

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 11 parameters and no output schema, the description carries the burden well: it states the return shape (answers plus mentioned brands and cited URLs), filtering dimensions, ordering, page size, truncation, and the full-text retrieval path. Nothing an agent needs to call this correctly appears to be missing.

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; the description goes further by explaining cross-parameter semantics — brand_mentioned=true/false combined with or without brand_name (defaulting to the workspace's own brand), and the full_text workflow requiring prompt_text/model/start_date/end_date to narrow down first. It mostly restates the per-field schema text, so it does not add a full extra level of meaning.

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 first sentence names the specific resource (the raw AI answers from ChatGPT, Perplexity, Gemini, Claude, Mistral, Google AI Overviews) and what each record carries (brands mentioned, URLs cited). It then explicitly separates this tool from the numeric siblings ('for numbers ... use those tools: they count every answer, this one returns a page'), so an agent can route without opening a 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?

Provides concrete trigger phrases ('what does ChatGPT say about us', 'show answers where we are not mentioned'), an explicit when-not-to-use rule pointing at the numeric tools, and an example query. It also covers the secondary use case of explaining a number another tool returned.

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