The Agent Skills Directory

[SAFE]: The skill operates entirely within the local file system, creating and reading markdown files for the purpose of career documentation. It does not perform network operations, execute shell commands, or access sensitive system directories beyond its own workspace folder.
[DATA_EXFILTRATION]: While the skill processes personal and professional data (achievements and company competency matrices), this information remains strictly on the user's local machine. There are no network calls, external API requests, or exfiltration patterns identified in the agents' instructions.
[PROMPT_INJECTION]: The skill ingests user-provided text for documentation purposes. While this constitutes an attack surface for indirect prompt injection, the skill possesses no capabilities (such as automated shell execution or network access) that would allow such an injection to be exploited. The instructions also emphasize user review and approval before any data is saved to disk.
[COMMAND_EXECUTION]: The skill's functionality is limited to prompting the user for information and managing markdown files. It does not use the shell to execute arbitrary commands or install software during operation.

brag-writer