Real-time analysis of your social media comments to identify emotional content.
Users
4
−1 in 30 days
Rating
0.0 / 5
unchanged in 30 days
Ratings
0
0 in 30 days
Store rank
#271K
#225K among extensions
Version
1.1
updated 6 Dec 2024
Users
monthly, 2 years
Latest, 15 Oct 2026:4 usersMove the mouse over the chart for any month
Ratings count
monthly, 2 years
Latest, 15 Oct 2026:0 ratingsMove the mouse over the chart for any month
The Chrome Web Store rounds user counts down to the leading digit, so the user line moves in steps. The ratings count is exact and is the better signal of real adoption.
About Empatia
Empatia provides real-time analysis of social media comments to identify emotional content and deliver inline feedback. Think of this feedback as a way to double check how you communicate on social media. Sometimes we write a comment that we'll regret writing, or regret writing the way we wrote it - I hope Empatia can help with that.
This project was primarily built as a learning experience. Suggestions for fixes and improvements are welcome.
Currently available to run on Reddit, BlueSky, and Twitter (X).
Getting Started
Select supported sites: Choose which sites (Reddit, Bluesky, or Twitter) Empatia will run on.
Adjust sensitivity: Set the sensitivity for emotional feedback (higher sensitivity will detect negative emotions with a lower threshold).
Set the minimum word count you want there to be before Empatia starts analyzing the text.
Write a comment or post: As you type, Empatia sends your text through sentiment analysis to assess its emotional polarity.
Positive feedback: Receive messages encouraging positive interactions if your comment is classified as positive.
Negative feedback: If your comment is negative, the system classifies the emotion and provides specific feedback (anger, annoyance, disgust, sadness, generic).
<a href="https://www.flaticon.com/free-icons/letter-e" title="letter e icons">Letter e icons created by arnikahossain - Flaticon</a>