That's solid advice.

I think combining the coding activity, the website, the whitepaper, and how the team interact with their community (on BitcoinTalk, Telegram, etc.) is a pretty good basis when it comes to evaluate the authenticity of a coin.

We want our analysis to stay neutral, but we might include an authenticity 'score' later on. We plan on developing our own ML models for various tasks if we successfully reach our fundraising goals such as: objectivity/subjectivity of social conversations, news authenticity, and more.

Even though social media might not be relevant to check the authenticity of a coin, it remains crucial in evaluating sentiment and overall reputation.