Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.stardeck.ai/llms.txt

Use this file to discover all available pages before exploring further.

The Stardeck agent can make cross-app calls on your behalf. Instead of writing SDK code yourself, you can ask the agent to call another app’s API, discover what endpoints are available, or orchestrate data flows across multiple apps.

​
What the Agent Can Do

​
From a Project Chat

When you’re chatting with the agent inside a specific project, it can:
  • Call another app’s API — “Call the inventory app’s /api/products endpoint and get the list of products”
  • Choose the environment — “Call the CRM app’s sandbox to test the new endpoint”
The agent uses your project’s deployment credentials, so it follows the same permission rules as your app’s code would.

​
From an Organization Chat

When chatting at the organization level, the agent has broader capabilities:
  • Call between any two apps — “Call the inventory app from the CRM app and create a new product”
  • Discover endpoints — “What API endpoints does the invoicing app expose?”
  • List apps — “Which apps in our org have cross-app calling enabled?”
The org-level agent can orchestrate calls between any pair of apps without needing to be “inside” either one.

​
Environment Selection

The agent can target either sandbox or production when making cross-app calls. By default, it calls the sandbox environment to keep things safe during development. You can specify which environment to use:
  • “Call the production endpoint on the billing app”
  • “Test the new route on the inventory app’s sandbox”

​
Restricting Agent Environments

You can control which environments the agent is allowed to call from each project. Go to the project’s Cross-App settings tab and configure Agent Callable Environments.
SettingWhat It Means
Sandbox onlyAgent can only call sandbox deployments (safe for development)
Production onlyAgent can only call production deployments
Both (default)Agent can call either environment
Restricting to sandbox-only prevents the agent from accidentally hitting production endpoints while you’re iterating.

​
Endpoint Discovery

Before making a call, the agent can discover what endpoints a target app exposes. This is especially useful when you don’t remember the exact path or want to see what data an endpoint expects. Ask the agent things like:
  • “What endpoints does the project-management app have?”
  • “Show me the API routes on the invoicing app”
  • “What does the /api/tasks endpoint expect as input?”
The agent reads the target app’s published endpoint list, including any documentation added alongside the API routes.

​
Permissions Still Apply

Agent calls go through the same permission checks as SDK calls. The agent cannot bypass grants:
  • The caller app needs a grant or allow-all access to the target app
  • Access level restrictions are enforced
  • Built-in rate limits apply
If a call is denied, the agent will tell you why and suggest setting up the appropriate grant.

​
Examples

Check inventory before creating an invoice:
“From the invoicing app, call the inventory app’s /api/products/SKU-123 to check stock, then create an invoice line item if it’s in stock.”
Sync data across apps:
“Get all customers from the CRM app and create corresponding contacts in the support app.”
Debug a cross-app integration:
“Call the payments app’s /api/webhooks/test endpoint from the storefront app’s sandbox and show me the response.”
Discover and explore:
“List all the apps in our org that accept cross-app calls, then show me what endpoints the top-selling app has.”