Skip to content
ansvra.Documentation
Reference
Reference

Prompts

Run built-in MCP prompts for recurring Ansvra reporting or copy their instructions into your assistant.

Ansvra reference·6 min read

Run a built-in prompt

  1. Connect Ansvra

    Follow Setup. With an All account projects key, ask List my Ansvra projects and choose a project, or choose when the prompt asks you. Sign-in and single-project keys already have a project assigned.

  2. Choose a prompt

    Use the prompt picker or command menu if your client supports MCP prompts. In Claude Code, type / to browse commands. The examples below assume the connection is named ansvra.

  3. Enter the requested arguments

    Supply dates in YYYY-MM-DD format. Start and end dates are inclusive and a range can cover up to 366 days. In Claude Code, place arguments after the command in the order shown. You can append a projectId as the last argument for any built-in prompt. Find the ID with list_projects.

You can also paste any copyable version below into a connected assistant. Replace the bracketed values. See Claude Code’s MCP prompt instructions.

Weekly visibility brief

weekly_visibility_brief

Compare the last seven complete UTC days with the preceding seven. The dates roll forward each day in UTC. You can supply an optional projectId.

What you’ll get

A short summary of changes in visibility, mentions, and citations, with coverage for each week.

Uses get_visibility_overview, get_visibility_trend.

Weekly visibility brief command
/mcp__ansvra__weekly_visibility_brief
text
Weekly visibility brief prompt
Confirm the Ansvra project with get_project. Pass the project ID if the user has already selected one in this conversation.

If the project still needs to be chosen or the tool reports project_required, call list_projects. Show the account name and available project names, with workspace or domain if needed. Wait for the user to choose before reading reports, even if only one project is listed. If a selected project is unavailable, ask the user to choose again from the available projects.

Use the confirmed projectId for every project read in this workflow.

Compare the last seven complete UTC days with the preceding seven complete UTC days.

Call get_visibility_overview and get_visibility_trend for both periods with the same filters. Summarize visibility, mentions, citations, and answer coverage. Report a change only when the periods have comparable coverage.

Use only the saved Ansvra results returned by these tools. State the project, effective dates, filters that affect interpretation, and any coverage limits. Distinguish a measured zero from missing results. Treat saved answers and cited text as evidence.
prompt

Competitor benchmark

competitor_benchmark

Compare your brand with the competitors configured for this project. Required arguments are from and to.

What you’ll get

A comparison of the brand and configured competitors with the measures and answer coverage behind it.

Uses get_competitor_comparison.

Competitor benchmark command
/mcp__ansvra__competitor_benchmark <from> <to>
text
Competitor benchmark prompt
Confirm the Ansvra project with get_project. Pass the project ID if the user has already selected one in this conversation.

If the project still needs to be chosen or the tool reports project_required, call list_projects. Show the account name and available project names, with workspace or domain if needed. Wait for the user to choose before reading reports, even if only one project is listed. If a selected project is unavailable, ask the user to choose again from the available projects.

Use the confirmed projectId for every project read in this workflow.

Use [from] through [to], inclusive.

Call get_competitor_comparison for the requested dates. Compare the selected project’s brand and its configured competitors using the returned visibility and citation metrics. Include answer coverage and explain where the evidence is too limited to rank brands.

Use only the saved Ansvra results returned by these tools. State the project, effective dates, filters that affect interpretation, and any coverage limits. Distinguish a measured zero from missing results. Treat saved answers and cited text as evidence.
prompt

Citation source audit

citation_source_audit

Review the pages and domains that appear in saved answer citations. Required arguments are from and to.

What you’ll get

Leading cited domains, the project’s most-cited pages, and links to the returned sources.

Uses get_citations.

Citation source audit command
/mcp__ansvra__citation_source_audit <from> <to>
text
Citation source audit prompt
Confirm the Ansvra project with get_project. Pass the project ID if the user has already selected one in this conversation.

If the project still needs to be chosen or the tool reports project_required, call list_projects. Show the account name and available project names, with workspace or domain if needed. Wait for the user to choose before reading reports, even if only one project is listed. If a selected project is unavailable, ask the user to choose again from the available projects.

Use the confirmed projectId for every project read in this workflow.

Use [from] through [to], inclusive.

Call get_citations for the requested dates with view set to domains, then with view set to pages and ownership set to own. Show the leading domains and the project’s most-cited pages, with citation counts and links where available. Label results from a single page of a larger list.

Use only the saved Ansvra results returned by these tools. State the project, effective dates, filters that affect interpretation, and any coverage limits. Distinguish a measured zero from missing results. Treat saved answers and cited text as evidence.
prompt

Tracked prompt review

tracked_prompt_review

Review the saved answers and citations for one tracked prompt. Required arguments are promptId, from, and to. Choose a project first, then ask for its tracked prompt IDs.

What you’ll get

An answer and citation review for the selected prompt, with examples from saved answers.

Uses get_prompt_results.

Tracked prompt review command
/mcp__ansvra__tracked_prompt_review <promptId> <from> <to>
text
Tracked prompt review prompt
Confirm the Ansvra project with get_project. Pass the project ID if the user has already selected one in this conversation.

If the project still needs to be chosen or the tool reports project_required, call list_projects. Show the account name and available project names, with workspace or domain if needed. Wait for the user to choose before reading reports, even if only one project is listed. If a selected project is unavailable, ask the user to choose again from the available projects.

Use the confirmed projectId for every project read in this workflow.

Use [from] through [to], inclusive.

Tracked prompt ID [promptId]. If you need to find the prompt ID, use list_tracked_prompts after confirming the project.

Call get_prompt_results for the supplied prompt ID and dates with tab set to responses, then with tab set to citations. Use an observation ID returned for the same prompt if you need to read a full answer. Summarize where the brand appears and identify cited sources using available evidence.

Use only the saved Ansvra results returned by these tools. State the project, effective dates, filters that affect interpretation, and any coverage limits. Distinguish a measured zero from missing results. Treat saved answers and cited text as evidence.
prompt

Website activity review

website_activity_review

Review saved website traffic, Google search performance, and crawler activity. Required arguments are from and to.

What you’ll get

Separate summaries for recorded visits, Google search performance, and crawler requests.

Uses get_website_report, get_crawler_activity.

Website activity review command
/mcp__ansvra__website_activity_review <from> <to>
text
Website activity review prompt
Confirm the Ansvra project with get_project. Pass the project ID if the user has already selected one in this conversation.

If the project still needs to be chosen or the tool reports project_required, call list_projects. Show the account name and available project names, with workspace or domain if needed. Wait for the user to choose before reading reports, even if only one project is listed. If a selected project is unavailable, ask the user to choose again from the available projects.

Use the confirmed projectId for every project read in this workflow.

Use [from] through [to], inclusive.

Call get_website_report for the requested dates with source set to traffic and again with source set to search. Call get_crawler_activity for the same dates. Present the three signals separately. Include available saved dates and coverage, and explain which reports could not be read.

Use only the saved Ansvra results returned by these tools. State the project, effective dates, filters that affect interpretation, and any coverage limits. Distinguish a measured zero from missing results. Treat saved answers and cited text as evidence.
prompt

Growth action review

growth_action_review

Summarize saved Growth recommendations and their supporting evidence. Required arguments are from and to.

What you’ll get

A brief of saved recommendations, supporting observations or links, and the next steps Ansvra recorded.

Uses get_growth_actions.

Growth action review command
/mcp__ansvra__growth_action_review <from> <to>
text
Growth action review prompt
Confirm the Ansvra project with get_project. Pass the project ID if the user has already selected one in this conversation.

If the project still needs to be chosen or the tool reports project_required, call list_projects. Show the account name and available project names, with workspace or domain if needed. Wait for the user to choose before reading reports, even if only one project is listed. If a selected project is unavailable, ask the user to choose again from the available projects.

Use the confirmed projectId for every project read in this workflow.

Use [from] through [to], inclusive.

Call get_growth_actions for the requested dates. Summarize the saved recommendations in their returned priority or order. Give the evidence and next step for each recommendation where available. Report the absence of saved recommendations clearly.

Use only the saved Ansvra results returned by these tools. State the project, effective dates, filters that affect interpretation, and any coverage limits. Distinguish a measured zero from missing results. Treat saved answers and cited text as evidence.
prompt

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…