Explicit Own-Credential Control

Agents can now be marked as requiring your own AI-provider credentials, making execution requirements clearer and preventing accidental fallback behavior.

This update made credential expectations more explicit.

Agents Can Require Your Own Provider Credentials

Some users want an agent to run only through credentials they control. Kindship now supports that as an agent-level setting instead of relying on implicit fallback behavior.

  • Agents can be marked as requiring your own provider credentials
  • Provisioning can inherit that preference from sibling agents
  • Blocked dispatches explain the credential requirement more clearly

Setup Failures Became Easier To See

Several provisioning and inner-loop update paths now fail more explicitly when an expected update did not actually apply.

That makes credential and setup problems easier to diagnose instead of turning them into quiet no-op states.