Use this when you need to know whose workspace this connection acts on and what it may do: the workspace name, plan tier and size (the contacts the plan holds), how it is billed, how many contacts count toward the limit right now, and the scopes the connection holds. Also use it when the founder asks about their plan, its limits or what it includes. Not for audience numbers in a report (use `get_audience_stats`) or for the projects (use `list_projects`). `complimentary` is true for a plan given rather than sold.
It also tells you the plan up front, read from the plan registry: `limits` (contact size, campaign emails a month, projects), `usage` against them (including `hidden_contacts`, saved contacts held past the size), and `includes` (whether the plan sends broadcasts, runs automations, saves custom segments, shows where signups came from, lists the page in search engines and removes the badge). Use it to mention a limit before the founder hits it, never to guess one. `upgrade_url` opens this workspace's plan page. When contacts are held there is also an `upgrade` object with the plan that holds them. Give the link when the founder asks about plans or something is blocked, and say which plan removes it and why it matters for what they asked. Never quote a price or a limit that this result does not state.
`key.mode` says how you are connected. `agent` means a signed-in person approved this app (OAuth, as in ChatGPT and Claude), so "me" is that person and tools that default to them (like `send_test_broadcast`) work without asking for an address. Any other mode is an API key with nobody behind it. `key.project` tells you whether the key is pinned to a single project, in which case every other tool can only reach that one.
`usage.contacts` is the BILLABLE count, which excludes test-mode and flagged contacts, so do not use it to report how many signups there are.