← All integrations

Your own agent

Anything that is not a browser can call your workspace.

A token lets your own code call the workspace YOUR SCRIPT OR AGENT Token Call YOUR WORKSPACE scan:run read a company from its website work:run a pitch, a call sheet, a push library:read a saved pitch, by name context:write an outcome, a next step ticked
You tick the scopes at mint. A token never gains one later.
Workspace settings in Prospektor, where the owner mints a token and ticks what it may do.
Workspace settings in Prospektor, where a token is minted.

A token lets something that is not a browser talk to your workspace: your CRM, an agent you are building, a script. Read a company, start a pitch, read a saved one, record an outcome, take a copy out.

You tick what a token may do when you mint it, and it never gains a scope later. It expires when you say. The secret is shown once; losing it costs one revoke and one mint.

Only the workspace's owner can mint one, because a token acts as the whole workspace with nobody watching. The card names each token, who minted it and when it expires.

  1. Open Connect your apps → Your own agent → New token in your workspace settings.
  2. Label it with what will hold it, tick what it may do, set when it expires, press Mint.
  3. Copy the secret now; nothing can show it again. Copy setup for your agent gives your own Claude the whole of it.

More integrations from Prospektor

See all integrations →