The Agent Skills Directory

[SAFE]: The skill is intended for academic and linguistic analysis, specifically focusing on Rhetorical Structure Theory (RST) and argument mining. All provided logic is transparent and utilizes standard Python libraries.
[COMMAND_EXECUTION]: The skill includes Python scripts designed to be run by the agent. These scripts use the re, collections, and statistics modules for text processing. There are no calls to os.system, subprocess, or other functions that could lead to arbitrary command execution on the host system.
[DATA_EXPOSURE]: The skill processes user-provided text corpora. While this involves reading potentially sensitive data provided by the user, the analysis is performed locally by the agent to generate a report, with no evidence of external data exfiltration or hardcoded credentials.
[PROMPT_INJECTION]: The skill is designed to analyze external text, which represents a surface for indirect prompt injection. However, the scripts focus on counting markers and measuring lengths using regular expressions, rather than executing or recursively processing the text as instructions, mitigating the risk of the agent following malicious commands embedded in the corpus.

rhetorical-discourse-structure