cyberneticlibrary

Analyze financial text sentiment

alphaear-sentimentskillsetup L2274
ninehills/skills
What it does

Analyze financial text sentiment and score

Best for

Scoring sentiment on bulk financial news when local FinBERT speed outweighs LLM latency.

Inputs
  • · text (financial article or news)
Outputs
  • · JSON: {score: float, label: str, reason: str}
Requires
  • · FinBERT (torch + transformers)
  • · SQLite3 (database)
Preconditions

FinBERT model downloaded; DatabaseManager initialized

Failure modes
  • · FinBERT model load timeout; missing database connection; invalid text format
Trust signals
  • · FinBERT model from transformers library (established)
  • · SQLite3 integration tested
  • · Score range [-1.0, 1.0] with labeled categories