# Implementation Plan: Event-Driven API Metrics Updates

## Phase 1: Event Infrastructure & Test Setup
Define the event mechanism and create baseline tests to ensure we don't break data accuracy.

- [ ] Task: Create `tests/test_api_events.py` to verify the new event emission logic in isolation.
- [ ] Task: Implement a simple `EventEmitter` or `Signal` class (if not already present) to handle decoupled communication.
- [ ] Task: Instrument `ai_client.py` with the event system, adding placeholders for the key lifecycle events.
- [ ] Task: Conductor - User Manual Verification 'Phase 1: Event Infrastructure & Test Setup' (Protocol in workflow.md)

## Phase 2: Client Instrumentation (API Lifecycle)
Update the AI client to emit events during actual API interactions.

- [ ] Task: Implement event emission for Gemini and Anthropic request/response cycles in `ai_client.py`.
- [ ] Task: Implement event emission for tool/function calls and stream processing.
- [ ] Task: Verify via tests that events carry the correct payload (token counts, session metadata).
- [ ] Task: Conductor - User Manual Verification 'Phase 2: Client Instrumentation (API Lifecycle)' (Protocol in workflow.md)

## Phase 3: GUI Integration & Decoupling
Connect the UI to the event system and remove polling logic.

- [ ] Task: Update `gui.py` to subscribe to API events and trigger metrics UI refreshes only upon event receipt.
- [ ] Task: Audit the `gui.py` render loop and remove all per-frame metrics calculations or display updates.
- [ ] Task: Verify that UI performance improves (reduced CPU/frame time) while metrics remain accurate.
- [ ] Task: Conductor - User Manual Verification 'Phase 3: GUI Integration & Decoupling' (Protocol in workflow.md)