The Agent Skills Directory

[SAFE]: The skill instructions and supporting documentation do not contain any malicious patterns, prompt injections, or obfuscated content. The logic is clearly explained and focused on the stated purpose of sentiment scoring.
[EXTERNAL_DOWNLOADS]: The documentation references well-known Python libraries such as vaderSentiment, transformers, and SnowNLP. It also cites established sentiment models hosted on Hugging Face, including ProsusAI/finbert and cardiffnlp/twitter-roberta-base-sentiment-latest. These are standard, reputable resources in the machine learning community and are documented neutrally.

algo-social-sentiment