> For the complete documentation index, see [llms.txt](https://sonic-ai-works.gitbook.io/sonic-networks-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://sonic-ai-works.gitbook.io/sonic-networks-docs/documentation/developers/ai-and-music/sonic-ai-v1.0-pro-control-plane.md).

# Sonic AI v1.0 Pro — Control Plane

Sonic AI v1.0 Pro control-plane implementation profile for model responsibility, registry routing, fallback, context, tools, generation operations and usage truth.

> **Implementation profile / early phase** — Documented routes and provider families describe intended control-plane behavior; they do not certify production availability.

Creators interact with **Sonic AI v1.0 Pro** through product-level modes such as `Auto`, `Pro`, `Fast`, `Efficient` or `Private`. Application code calls SONIC abstractions instead of scattering provider IDs, credentials and provider-specific schemas through clients.

## Responsibility separation

```
Creative intelligence
  Sonic AI routing
  OpenAI / other reasoning provider where configured
  prompt enhancement / analysis / structured output

Music execution
  configured music-provider adapter
  current MVP: primarily Suno
  produces audio-provider job/results
```

A reasoning-provider logo or selection must never imply that provider generated the audio unless it actually executed the music-generation operation.

## Control-plane path

```
Normalized Sonic AI Request
→ server authority + workspace policy
→ intent + capabilities
→ permission-filtered context
→ responsibility classification
→ Model Registry + Router
→ qualified provider/model
→ SONIC tool / model execution
→ validation
→ artifact / generation brief / domain result
→ normalized usage
```

Music execution then passes through the separate durable generation boundary.

## Stable SONIC boundary

Public/internal product contracts use SONIC concepts: profile, task intent, capability, context policy, routing policy, responsibility, tool, artifact, operation and usage. Provider-specific IDs remain server registry data.

## Routing and fallback

Eligibility is evaluated before ranking. Capability, responsibility, privacy, environment, health, workspace policy and provider policy are hard filters. Quality, latency and cost telemetry rank only eligible routes.

Gateway retry and semantic-model fallback are separate layers. A creator manual override fails transparently unless fallback is explicitly allowed. Streaming never silently switches semantic models after content has begun.

## Context and tools

Context retrieval is task-scoped and permission-filtered. Tools declare permissions, mutation behavior and confirmation requirements. Authorization is re-checked at tool execution.

Sonic AI may prepare a draft or transaction intent but cannot sign a user's wallet, allocate unauthorized rewards or bypass settlement policy.

## Artifacts

```
DRAFT → APPROVED → APPLIED
             └────→ REJECTED
```

Artifacts retain originating operation and provenance. A refined music brief is an artifact/input transformation; it is not proof of audio generation.

## Usage truth

Record what actually executed:

```
Sonic profile / mode
task
responsibility
provider / providerModelId
routingReason
transport
usage units where available
latency
terminalStatus
```

For generation workflows, AI Credit reservation/settlement belongs to the durable GenerationOperation rather than a client-estimated provider cost.

## Generation boundary

Music providers implement a generic adapter. `GenerationOperation` remains the SONIC business identity.

```
request
→ validate
→ create/recover idempotent operation
→ reserve credits
→ dispatch selected music provider
→ callback / poll merge
→ validate normalized result
→ completed | failed | cancelled
→ settle | release
```

Dispatch requires valid server authority, schema, URL policy, configured provider eligibility, atomic credit cost, an existing operation and successful reservation.

Callbacks merge by stable track identity and monotonic progression. Ledger retries never redispatch provider work.

## Registry API rule

A safe creator-facing model response may expose branded labels, capability summaries and availability, but not private provider credentials, privileged model configuration, internal cost rules or routing secrets.

Recommended public semantics:

```
id
label
providerBrand
responsibility
capabilities
availability
modeCompatibility
```

## Secrets

Provider credentials are server-only. Never expose them through client environment variables, public source maps, browser network payloads or provider error forwarding.

## Control-plane invariant

```
SONIC abstraction
+ explicit responsibility classification
+ server registry
+ qualified routing
+ permission-filtered context
+ execution-time authorization
+ durable generation boundary
+ normalized usage truth
= Sonic AI v1.0 Pro control plane
```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://sonic-ai-works.gitbook.io/sonic-networks-docs/documentation/developers/ai-and-music/sonic-ai-v1.0-pro-control-plane.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
