The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill performs deterministic text analysis.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted text (seed body) to identify linguistic markers. The ingestion point is the 'seed body' mentioned in the workflow. The skill uses concrete regex-based detection rules rather than open-ended LLM interpretation, and it lacks dangerous capabilities such as file writing, network operations, or subprocess execution, minimizing the risk of the input text influencing agent behavior.
[DATA_EXPOSURE]: The skill only interacts with non-sensitive files such as a local style guide and input text. No credentials or sensitive system files (e.g., .ssh, .aws) are accessed.

score-intuition-density