AGENT 26 / 11 · RETAIL & E-COMMERCE · WEALTH & ASSET MANAGEMENT

Market Sentiment Aggregator

Blends news and social tone, earnings call transcript sentiment and alternative data into a single research signal, with mention volume amplifying conviction.

READS FROMNews/social feeds, earnings transcripts, alternative data

MCP SERVERWealth Data Server

MODELRouted to reasoning-tier open model

HANDS OFF TORobo-Advisory Profiling Agent

RUN THE AGENT

Set the source signals, generate the composite score

Adjust news tone, earnings sentiment and alt-data signal and the agent recalculates the blended sentiment score instantly.

ASSUMPTIONS

Daily news / social mentions 400

Positive sentiment share 48%

Negative sentiment share 22%

Latest earnings call sentiment 15

Alternative data signal 8

Weight given to mention volume 30%

—

Composite sentiment score

—

Signal tier

—

News sentiment score

—

Volume-adjusted signal

Metric	Value

HOW IT WORKS

What the agent actually does

1. Ingest multi-source signals	Pulls news/social mention volume and tone, earnings call transcript sentiment, and alternative data feeds — web traffic, card spend, app downloads — for the name.
2. Score each source	Converts each source into a standardized sentiment score on a common scale.
3. Blend into a composite	Weights and combines the sources into a single signal, with mention volume amplifying conviction.
4. Hand off	Surfaces the signal to the Robo-Advisory Profiling Agent and portfolio managers as a research input, and flags sharp sentiment shifts for immediate review.