get_monthly_accuracy_trend
Purpose: Monthly accuracy time series per (category, target_market, lag_bucket). Use to verify sustained performance and detect recent degradation. Triggers (casual questions too): "is accuracy improving?", "적중률이 좋아지고 있어?", "monthly performance trend?", "최근에 예측 성능 떨어졌어?", "show accuracy over time". When to call: after get_prediction_accuracy and get_backtest_tuning_state — completes the trust chain. Prerequisites: get_prediction_accuracy recommended. Next steps: none (trust chain complete). Caveats: excludes the 'all' month aggregate; empty when backtest_results is unpopulated.
Args: category: Optional category filter target_market: Optional target market filter
Disclaimer: Information only, not investment advice.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| category | No | ||
| target_market | No |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
| error | No | Set true on error responses | |
| action | No | Recommended client action (error path) | |
| reason | No | Human-readable cause (error path) | |
| full_data | No | ||
| retryable | No | Whether the client should retry (error path) | |
| timestamp | Yes | RFC3339 UTC, server build time | |
| ai_summary | No | One-line AI-oriented summary (success path) | |
| disclaimer | Yes | Canonical compliance disclaimer (always present) | |
| error_code | No | Stable error identifier; see mcp_error_policy.md | |
| request_id | Yes | 32-hex per-response correlation id | |
| _llm_summary | No | ||
| action_value | No | ||
| _next_actions | No | ||
| fallback_note | No | ||
| fallback_tool | No | Suggested fallback (error path) | |
| is_real_money | No | ||
| _value_signals | No | ||
| summary_for_user | No | One-line jargon-free Korean summary (success path) | |
| data_classification | No | ||
| is_investment_advice | No | ||
| ai_summary_ttl_seconds | No | ||
| _market_state_narrative | No | ||
| ai_summary_generated_at | No | RFC3339 UTC | |
| _followup_questions_for_user | No |