AI Sentiment MCP tool

cite42_sentimentAbout AI Sentiment

Score how each AI model talks about a brand (positive, neutral or negative) and the verbatim phrases driving the tone, straight from your agent.

Overview

Not just whether the models mention a brand, but how. Your agent gets a positive/neutral/negative split per model plus the phrases driving it, and an aggregate across all models queried.

Reading the metrics

Sentiment percentages are AI-estimated tone weights, not the share of positive answers. Failed classifications are excluded; no usable answers means unavailable, not 0%.

Example

# you say
How do the AI models talk about Acme CRM for startup teams? Is the sentiment positive?

# your AI replies
Mostly positive: 62% positive, just 8% negative across the models. They reach for phrases like “quick setup for small teams” and “easy to learn”. The main reservation is reporting depth for large teams, so that’s the narrative to get ahead of.

View tool JSON

{
  "brand": "Acme CRM",
  "perModel": [
    { "model": "chatgpt", "positive": 0.55, "phrases": ["quick setup for small teams"] }
  ],
  "aggregate": { "positive": 0.62, "negative": 0.08 },
  "billing": { "cost": { "usd": "0.30" } }
}

Full schema: POST /v1/sentiment.