Skip to main content

Management

Management tools let an AI assistant find, read, create and maintain your content directly over MCP — so it can discover a prompt, revise it, or tidy up, without you switching to the dashboard.

Today this covers prompts, resources, and custom tools:

Plus one tool the agent calls first to learn where it can work:

  • bootstrap — report the agent's scope and the teams it can manage

Finding and reading content​

Before it can revise or copy something, an agent has to find it and read it. The list_* tools enumerate what you can see — your own content, your teams', and your organization's shared catalogue — and the get_* tools return one item's full detail: a prompt's template, a resource's content, or a custom tool's connector definition.

What you can see is deliberately broader than what you can change. An agent can read a prompt from the shared catalogue and copy it into its own space to adapt, even though editing the catalogue itself stays a dashboard action. Anything you can't see is reported as not found, so nothing about its existence leaks. (For custom tools, only enabled ones are listed — matching what an agent can actually call; one still awaiting its secret stays out of the listing.)

Where content is created​

You never pass an owner or scope. Everything you create lands in your own space by default; to target a team you belong to, you name that team. Prompts and custom tools spell this one way, resources another, because a resource's path can itself contain /:

  • Prompts and custom tools are addressed by name, with an optional <team>/ prefix — launch-copy (yours) vs marketing/launch-copy (the marketing team).
  • Resources are addressed by their path, with an optional separate team field — docs/readme.md (yours) vs docs/readme.md with team: "marketing" (the marketing team).

Either way, you can only target a team you belong to and are allowed to manage. The agent learns which team names it can use by calling bootstrap first — an empty team won't show up in a listing, so this is the only reliable way to discover it.

There is no way to publish to your organization's shared catalogue from here — that stays a deliberate action you take in the dashboard.

Custom-tool secrets stay in the dashboard​

A custom tool that calls an authenticated API needs a secret (a bearer token, an API key, a password). For safety, secrets are never set over MCP — you define the tool and its auth type, and a person adds the secret in the dashboard. Until then the tool is created disabled and can't run. Tools that need no authentication work end to end without a human step.

Permissions​

Every call obeys the same permissions your credential carries — exactly what you'd need to do the same thing in the dashboard. If your API key or authorization can't create the content, or can't manage the team you addressed, the tool is refused rather than silently doing nothing. This is how you keep an agent scoped: grant it only what you want it to change.

Like any built-in tool, a management tool also has to be available to your credential (see Discovering tools); an organization can choose which tools its agents may use. A credential that isn't allowed to use them never sees them — they are omitted from tools/list and a direct call is rejected as an unknown method.

Metering​

A management call is a tool call like any other: it counts once against your plan's tool-usage allowance and appears in your usage history. Refused calls (for example, a missing permission) are counted too.

Learn more​

This page is the tool reference. For the bigger picture and hands-on walkthroughs: