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
What remains different behind a service provider?
Permissions, latency and side effects remain part of the service contract.
A common entry point with real remote constraints
An HTTP, MCP or messaging provider can organize external resources and actions in AFS. It adapts paths, arguments and results to a real service.
Reading a message and sending one remain different operations. Reads may paginate or hit rate limits. Sends may communicate with another person. A timeout does not necessarily mean nothing happened; understand idempotency or deduplication before retrying a send.
What should the provider explain?
- Resource identity, including references across pagination.
- Arguments, results and possible external effects.
- Authentication failure, denial, missing resources, throttling and temporary failures.
- Cached state and ways to discover updates.
Turning every failure into an empty list makes “nothing exists” indistinguishable from “the read failed.” A common interface should help consumers discover these differences consistently.
Check your understanding
What remains different behind a service provider?
Permissions, latency and side effects remain part of the service contract.
Continue along a learning path
- Tour the providersStep 5 of 5