finance-sentiment - Cross-Platform Stock Sentiment Analysis Skill

Skill Overview

finance-sentiment is a read-only stock sentiment research skill that uses the Adanos Finance API to aggregate data from four sources—Reddit, X.com, news, and Polymarket—in a single request. It transforms scattered discussions across social platforms into structured signals such as mention volume, buzz score, bullish percentage, and trends, providing a quick and comparable answer to the question, “What is the market talking about?”

Use Cases

1. Query the social discussion volume for a single stock: When you want to know “How many people are discussing TSLA on Reddit?” or “How much buzz does NVDA have on X.com?”, a single request provides the mention count, buzz score, and bullish percentage.
2. Compare multiple stocks side by side: Include up to 10 stocks such as AMD, NVDA, and META in the same request, rank them by buzz score or bullish percentage, and quickly identify which stock is currently receiving more attention and where the largest bullish-bearish divergence exists.
3. Verify cross-platform opinion consistency: Retrieve sentiment data from Reddit and X.com separately, compare their bullish percentages and trends, and determine whether retail communities and X users are “uniformly bullish,” “uniformly bearish,” or “clearly divided.”

Core Features

1. Four-source sentiment aggregation: Covers four source categories: Reddit, X.com, financial news, and Polymarket. Social and news sources provide mentions, while Polymarket provides the number of trades (trade_count). Missing sources are explicitly marked as “no data” rather than incorrectly interpreted as neutral.
2. Standardized sentiment indicators: Each source outputs a consistent set of fields—buzz score (maximum 100), bullish percentage (bullish_pct), and trend direction (such as rising). The default lookback period is 7 days, with custom time windows also supported, enabling side-by-side comparisons across stocks and platforms.
3. Batch comparison and read-only safety boundary: The compare endpoint supports batch queries for 1–10 tickers at a time, making it compact and suitable for rapid research. The skill is entirely read-only: it only retrieves and analyzes data and does not execute trades or perform any modification operations.

Frequently Asked Questions

Can finance-sentiment place trades for me?

No. This is a read-only skill that can only query and organize stock sentiment data from Reddit, X.com, news, and Polymarket. It does not execute trades or generate trading instructions. All outputs are research reference signals and do not constitute investment advice.

What prerequisites are required to use this skill?

You need to set the ADANOS_API_KEY environment variable (you can configure it with export ADANOS_API_KEY="sk_live_..."). All requests are authenticated using the X-API-Key request header. API documentation is available at api.adanos.org/docs.

What do bullish percentage (bullish_pct) and buzz score (buzz_score) mean?

The bullish percentage represents the proportion of content from a given source expressing a bullish view. For example, 31% means that approximately three in ten discussions are bullish. The buzz score is a 0–100 discussion activity rating that measures “how many people are talking about this stock.” The former reflects the direction of opinions, while the latter reflects the level of attention. Both are needed to fully characterize a stock’s social sentiment.

How many stocks can be compared at once?

The compare endpoint supports passing 1 to 10 tickers in a single request. For example, tickers=AMD,NVDA,META retrieves sentiment data for all three stocks from the specified sources in one request, making it suitable for batch comparisons.