Skip to main content
Ask your agent this, with <product-name> replaced by your product’s name:
The agent answers with numbers from TPC’s ChatGPT plugin discovery measurement for that product.

​
Before you ask

  • Connect the server and sign in. See Connect the MCP server.
  • If you belong to more than one organization, put your organization slug in the same message. See Find your organization slug.
  • Use the product name as it appears in TPC. The agent searches for it.
  • Your organization needs an Enterprise plan.
  • The product needs a ChatGPT plugin ID. If it has none, ask the agent to set it. Changing the plugin ID needs an owner or admin role.
  • The measurement uses the prompts in the product’s plugin discovery lane. These are separate from the product’s tracked prompts. If the lane is empty, ask the agent to add some.

​
What the agent does

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

Find the product

If you did not give your organization slug, list_organizations finds it. Then list_products with your organizationSlug and the product name as query returns the product’s slug and ID, which the next calls need.
2

Get the headline numbers

get_plugin_discovery with no dimension returns the summary: measured and search runs, shortlisted, suggested, and card shown runs, the plugin’s average rank in the search results, and the plugins ChatGPT picked instead. A product with no plugin ID set returns optedIn: false.
3

See the trend

get_plugin_discovery with dimension: "date" returns one row per UTC day from the first measured run through today, with that day’s runs and rolling 30-day sums. It is not limited to 30 rows. A product with no plugin ID set or no runs returns { days: [] }.
4

Compare with other plugins

get_plugin_discovery_ranking returns the top 10 plugins ChatGPT suggested, ranked by suggested runs. If your plugin is outside the top 10, it is returned separately.
5

Look at the prompts behind it

list_plugin_discovery_prompts returns each prompt in the lane with how often it has run and its last run.
A run is one ChatGPT answer to a plugin discovery prompt. A plugin is shortlisted when it appears in the search results ChatGPT considered, suggested when ChatGPT recommends it in the answer, and card shown when ChatGPT renders its install card.

​
What you get back

The agent writes a summary from those calls. Depending on which calls it makes, the summary can include:
  • Measured, search, shortlisted, suggested, and card shown runs, and your plugin’s average rank
  • The plugins ChatGPT picked instead of yours
  • A daily trend with rolling 30-day sums
  • The top 10 plugins by suggested runs, and where yours stands if it is outside them
  • The prompts in the lane, with runs, suggested runs, and last run

​
Change what is measured

Ask the agent to change the lane or the plugin ID. These tools write, so the agent asks for your go-ahead first. Adding text that is already in the lane returns the existing prompt. Saving the same plugin ID again keeps the original opt-in date, and a different ID restarts it. The same data is available from the terminal. See ChatGPT plugin discovery in the CLI reference.