The Agent Skills Directory

[SAFE]: The skill manifest and instructions describe a utility for social media account management via browser automation. No evidence of malicious intent or hidden capabilities was found.
[COMMAND_EXECUTION]: The skill describes the use of JavaScript via the browser's developer console. While this involves code execution, it is presented as a manual user action for legitimate account cleanup, and no automated or remote command execution patterns are present.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates processing of external data (X/Twitter search results). While this constitutes an attack surface for indirect prompt injection if the agent were to process these results directly, the current implementation provides scripts for the user to run manually, posing minimal risk to the agent environment.

content-cleanup