The Agent Skills Directory

[SAFE]: No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected. The term 'exploit' in the description refers to the decision-making phase of writing (using existing materials) rather than a security vulnerability.
[COMMAND_EXECUTION]: The skill performs file system operations (read/write) on paths specified by the user to save and retrieve article progress. This is the intended primary purpose of the skill to maintain a 'choose-your-own-adventure' writing workflow.
[DATA_EXPOSURE]: The skill processes user-provided markdown files ('raw materials'). While it interacts with external data, it does so within the context of text generation and does not attempt to access sensitive system files or credentials.

writing-beats