You are here. See how this question connects to other ideas.
Select a node to open its page · Expand to read within the map
One question
How should you navigate the providers?
Classify resources and responsibilities before repository directories.
Choose by the work you need
| Work | Representative implementations | First questions |
|---|---|---|
| Access stored data | FS, JSON, KV, DID Space | Ownership, persistence, read/write capabilities |
| Inspect a running environment | proc, registry, observability, UI | Lifetime, caller scope, operation effects |
| Connect an external service | HTTP, MCP, messaging providers | Authentication, limits, remote failures and side effects |
These are learning groups, not mutually exclusive types. Repository directories such as core, basic and runtime organize implementations. They do not prove that every deployment mounts those providers.
A catalog is not live discovery
To read a task, discover actual mounts in the current workspace, then inspect the path and its capabilities. Source code establishes an implementation to study; it does not establish configured credentials or anonymous access.
Start with JSON for structural mapping, FS for a local boundary, and MCP for protocol adaptation. Ask the same questions of each: how names resolve, what is returned, what operations do and how failures are expressed.
Continue with the provider catalog, an implementation guide rather than a snapshot of your session.
Check your understanding
How should you navigate the providers?
Classify resources and responsibilities before repository directories.
Continue along a learning path
- Tour the providersStep 2 of 5