Specification: OpenAI Provider Integration

Overview

This track introduces support for OpenAI as a first-class model provider. It involves implementing a dedicated client in src/ai_client.py, updating configuration models, enhancing the GUI for provider selection, and integrating OpenAI into the tiered MMA architecture.

Functional Requirements

1. Core AI Client (`src/ai_client.py`)

OpenAI Integration:
- Implement _send_openai() to handle communication with OpenAI's Chat Completions API.
- Implement a tool-call loop similar to _send_gemini and _send_anthropic.
- Support for function calling (using tools and tool_choice).
- Support for multi-modal input (Vision) by encoding screenshots as base64 data URIs in the message payload.
- Implement response streaming support compatible with the existing GUI mechanism.
State Management:
- Define module-level _openai_client and _openai_history.
- Ensure ai_client.reset_session() clears OpenAI-specific history.
Error Handling:
- Implement _classify_openai_error() to map OpenAI API exceptions to ProviderError types (quota, rate_limit, auth, balance, network).
Model Discovery:
- Implement _list_openai_models() to fetch available models from the API.

2. Configuration & Authentication

Credentials:
- Update _load_credentials() to support an [openai] section in credentials.toml.
- Update the FileNotFoundError message in _load_credentials() to include the OpenAI example.
Cost Tracking:
- Update src/cost_tracker.py with pricing for gpt-4o, gpt-4o-mini, o1, and o3-mini.

3. GUI & Controller Integration

Provider Lists:
- Add openai to the PROVIDERS list in src/gui_2.py and src/app_controller.py.
Model Fetching:
- Ensure AppController._fetch_models() correctly dispatches to ai_client.list_models("openai").
Settings UI:
- The AI Settings panel should automatically handle the new provider and its models once added to the lists.

4. MMA Orchestration

Tier Support:
- Verify that agents in all tiers (1-4) can be configured to use OpenAI models via the mma_tier_usage dict in AppController.
- Ensure run_worker_lifecycle in src/multi_agent_conductor.py correctly passes OpenAI model names to ai_client.send().

Non-Functional Requirements

Consistency: Follow the established pattern of module-level singleton state in ai_client.py.
Latency: Ensure the tool-call loop overhead is minimal.
Security: Rigorously protect the OpenAI API key; ensure it is never logged or exposed in the GUI.

Acceptance Criteria

OpenAI can be selected as a provider in the AI Settings panel.
Models like gpt-4o and gpt-4o-mini are listed and selectable.
Agents can successfully use tools (e.g., read_file, run_powershell) using OpenAI models.
Screenshots are correctly processed and described by vision-capable OpenAI models.
Response streaming is functional in the Discussion panel.
Estimated costs for OpenAI calls are displayed in the MMA Dashboard.

Out of Scope

Support for the legacy gpt-3.5-turbo-instruct or other non-chat models.
Batch API support.
Fine-tuning management.

3.2 KiB Raw Blame History