Skip to content
ansvra.Documentation
Reference
Reference

Tools

Find available projects and see the inputs, defaults, and saved data for each Ansvra MCP tool.

Ansvra reference·Reference

Select a project

Call list_projects to find the issuing account and projects this connection can read. With an All account projects key, choose a project and pass its projectId to every project tool, even when only one project is listed. A call without it returns project_required.

With sign-in or a single-project key, you can omit projectId to use the assigned project. An ID outside the connection’s access returns not_found. Project selection applies to one tool call at a time.

Read tool results

Each successful tool call includes data and meta.requestId. Project reads also include meta.projectId and meta.resultScope set to saved. Reports include the effective scope or range where applicable. Look at both the result values and their coverage before reporting a zero or comparing periods.

The examples show tool arguments. Replace example dates and IDs with values for your project. Include projectId in each project read when using an all-projects key. If a result is too large, shorten the period or use the tool’s pagination.

Shared brand report filters

from and to are optional together. The brand reports use a 28-day window by default. Dates use YYYY-MM-DD and cover up to 366 days inclusive. A project’s reporting start date can shorten the effective window.

InputWhat it does
platformsOptional list of platforms. Omit it for all eligible saved results in scope.
promptIds and categoryIdsOptional lists of IDs from the connected project, up to 30 in each list.
market and languageOptional reporting filters. Omit them to use the project’s defaults.
promptTypeDefaults to nonbranded. Use branded for prompts naming the brand or all for both groups.

list_projects

Find the issuing account and projects available through the connection.

Inputs

limit defaults to 20 and accepts 1–50. Pass pagination.nextCursor as cursor with the same limit to read the next page.

Returns

Active project IDs, names, domains, and workspace IDs and names. meta.account contains the account ID and name. meta.projectAccess is project for an assigned project or account for an all-projects key. An all-projects key can return an empty list when there are no eligible projects.

list_projects arguments
{
  "limit": 20
}
json

get_project

Confirm the selected project and read its details.

Inputs

Pass a projectId from list_projects. An empty object uses the assigned project for sign-in and single-project keys.

Returns

Project ID, name, domain, website, market, and language, where recorded.

get_project arguments
{
  "projectId": "your-project-id"
}
json

list_tracked_prompts

Find prompts and the IDs needed to review their saved answers.

Inputs

limit defaults to 20 and accepts 1–50. Optional filters are active, archived, branded, market, and language. Archived prompts are excluded by default. Pass pagination.nextCursor as cursor and keep the same filters and limit for the next page.

Returns

Prompt IDs, text, category IDs when assigned, market, language, active status, and pagination.

list_tracked_prompts arguments
{
  "active": true,
  "limit": 20
}
json

get_prompt_results

Review saved results for a prompt or read a full saved answer.

Inputs

promptId is required. Optional inputs are from, to, platforms, tab, observationId, and page. Tabs are overview (default), responses, citations, and fanout when search queries are available. Read a full answer by passing its observation ID. The default date range is 28 days. Paged views start at page 1.

Returns

Prompt details and reporting scope, with the selected overview, responses, citations, or search queries. A full-answer request returns the matching response when it is in scope.

get_prompt_results arguments
{
  "promptId": "your-prompt-id",
  "from": "2026-10-01",
  "to": "2026-10-07",
  "tab": "responses"
}
json

get_visibility_overview

Read visibility, mentions, citations, and coverage for a selected period.

Inputs

Uses the brand report filters.

Returns

Brand measures, usable answer and prompt counts, comparisons when available, and reporting scope.

get_visibility_overview arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07",
  "promptType": "all"
}
json

get_visibility_trend

Read the time series behind changes in visibility and citations.

Inputs

Uses the brand report filters.

Returns

Date buckets with saved brand measures and the effective reporting scope.

get_visibility_trend arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07"
}
json

get_competitor_comparison

Compare the project’s brand with its configured competitors when competitor reporting is available.

Inputs

Uses the brand report filters.

Returns

Brand rows, period comparisons, and pairwise counts of which brands appeared in the same answers, when available.

get_competitor_comparison arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07",
  "promptType": "nonbranded"
}
json

get_brand_sentiment

Read saved sentiment and the coverage of sentiment classifications.

Inputs

Uses the brand report filters.

Returns

Tone counts, classification coverage, and available breakdowns or themes. Use coverage when interpreting changes.

get_brand_sentiment arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07"
}
json

get_shopping_visibility

Review saved product appearances and shopping results within AI answers.

Inputs

Uses the brand report filters.

Returns

Shopping coverage and available product, brand, retailer, placement, and trend summaries.

get_shopping_visibility arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07"
}
json

get_citations

Find cited pages or domains in saved answers.

Inputs

from and to default to a 28-day reporting window. Optional filters are platforms, market, language, domain, and ownership (all, own, competitor, other, or unclassified). view is pages by default or domains. page defaults to 1 and each page contains up to 50 rows.

Returns

Citation totals and paged source rows with counts and reporting context.

get_citations arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07",
  "view": "pages",
  "ownership": "own",
  "page": 1
}
json

get_website_report

Read a saved website report and its coverage.

Inputs

source is required and accepts traffic or search. Supply from and to together to request a date range. The default range is 28 days.

Returns

Saved website traffic or search performance with available coverage. Read meta.savedAt and meta.availability to establish whether a matching report is available.

get_website_report arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07",
  "source": "search"
}
json

get_crawler_activity

Read recorded crawler activity and see the busiest pages and bots.

Inputs

from and to default to a 30-day reporting window. Optional platform or bot narrows the report. view accepts overview (default), pages, or bots. page defaults to 1. Use pagination.hasMore to see whether more page or bot rows are available.

Returns

Recorded request totals, trends, coverage, and paged sets of up to 25 page and bot rows each.

get_crawler_activity arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07",
  "view": "bots",
  "page": 1
}
json

get_growth_actions

Read saved recommendations and the evidence behind them.

Inputs

Uses the brand report filters.

Returns

Saved Growth actions, reviews, work in progress when present, and available evidence or next review details.

get_growth_actions arguments
{
  "from": "2026-10-01",
  "to": "2026-10-07"
}
json

Ansvra

A clearer next step.

MEET YOUR AI GUIDE

A little clarity.
A clearer next move.

Ask about Ansvra, AI visibility, or your next step with content.

Opening chat…