Provider Sync Overview

Provider sync lets OAT manage one canonical set of agent assets and then project those assets into provider-specific views. In practice, that means you edit the canonical layout in .agents/ and .oat/, then let OAT generate or reconcile the provider copies.

This is useful when you want consistent instructions, skills, and rules across multiple providers without treating each provider directory as a long-lived hand-maintained source of truth.

What OAT Treats As Canonical

• canonical skills, agents, and rules under .agents/
• sync state and related metadata under .oat/
• provider-specific files as derived views unless they are explicitly adopted back into canonical form

Typical Flow

1. Run oat init to create the base OAT layout and setup state.
2. Inspect current sync state with oat status.
3. Adjust provider enablement with oat providers ... if needed.
4. Run oat sync to materialize provider views from canonical assets.
5. Re-run oat status after edits to confirm whether anything drifted or needs adoption.

When To Use This Section

Use Provider Sync when:

• you want to adopt OAT incrementally, without committing to tracked project workflows yet
• you need a clear model for drift, strays, and adoption
• you want to understand which files should be edited directly and which should be regenerated

Continue Here

• CLI Bootstrap for the first sync setup flow
• Scope and Surface for the canonical/provider-view model
• Commands for the exact CLI behavior