Auto-generated. This article is rebuilt from FastAPI route introspection by scripts/build_api_kb.py. Edit the route docstring in app/ and re-run the script — do not edit this file directly.

GET /api/v1/sentiment-comparison

Handler: app.signals.routes.get_sentiment_comparison
Methods: GET

Description

Per-article FinRep-AI news sentiment over the most-recent rows.

Returns the most-recent limit rows from news_items with the FinRep-AI sentiment reading attached. The endpoint name and shape are retained from the D6 parallel-run era; the AlphaVantage comparison leg retired 2026-06 per DOCTRINE.md D22 (FinRep-AI is now the sole sentiment source), so aggregated_av is permanently null.

Notes:

• finrep_ai_score / finrep_ai_label / finrep_ai_run_at / finrep_ai_model are populated when the FinRep-AI module successfully classified the row. finrep_ai_score is in [-1, +1] where positive means bullish and negative means bearish. finrep_ai_label is one of positive / negative / neutral. finrep_ai_model records which Claude model produced the row; pre-pivot rows scored by the legacy CPU FinBERT setup carry NULL there. Either score/label column being NULL means “FinRep-AI was unavailable when this row first landed” and the row is surfaced with finrep_ai_label = "unavailable".
• aggregated_av is always null (AlphaVantage retired). The companion aggregated_av_note carries the retirement reason per DOCTRINE P0 (no silent degradation).

Returns: {count, limit, items: [{idempotency_key, source, title, published_at, finrep_ai_score, finrep_ai_label, finrep_ai_run_at, finrep_ai_model}], aggregated_av: null, aggregated_av_note: str}

Parameters

Curl

curl -s https://bigclawd.com/api/v1/sentiment-comparison