Skip to main content
Ask your agent this, with X replaced by your product’s name:
The agent answers with numbers from TPC analytics for that product.

​
Before you ask

  • Connect the server and sign in. See Connect the MCP server.
  • Have your organization slug ready and put it in the same message. See Find your organization slug.
  • Use the product name as it appears in TPC. The agent searches for it.
  • Name a date range if you want one. Without one, the date window is the last 30 days.
TPC measures share of voice on the prompts it tracks for the product. See Add prompts.

​
What the agent does

The agent chooses its own calls. A typical answer uses these tools in this order.
1

Find the product

list-products with your organizationSlug and the product name as query. It returns the product’s slug and ID, which the next calls need.
2

Get the headline numbers

explore-analytics with metric: "sov" and no dimension. It returns your share of voice (sov), mentions, and runs from the latest snapshot in the date window.
3

Break the numbers down

explore-analytics with metric: "sov" and one dimension. Use competitor for your rank among competitors, engine for each answer engine, and prompt or topic for share of voice on each prompt or topic.
4

Add context

explore-analytics with metric: "citations" and dimension: "source" for the sources answer engines cite, metric: "sentiment" for positive, neutral, and negative mentions per prompt, and metric: "traffic" for visits from AI crawlers and assistants.
Each explore-analytics call returns one metric with at most one dimension, as columns and rows, along with the product and date window it used.
Without a dimension, or with dimension: "date", share of voice, mentions, and runs are trailing 30-day snapshot values. With dimension: "date", each row is a separate snapshot, so do not add up mentions or runs across rows.

​
What you get back

The agent writes a summary from those calls. Depending on which calls it makes, the summary can include:
  • Your share of voice, mentions, and runs
  • Your rank among competitors
  • Share of voice for each answer engine, prompt, or topic
  • The sources answer engines cite, with citation counts
  • Positive, neutral, and negative mentions per prompt, with reasons for the negative ones and an optional week-over-week change
  • AI visits in total, or by date, page, model, or category
See Metrics for what each number means. To act on the results, see Help me improve the AEO of my product X.