The Agent Skills Directory

[DATA_EXFILTRATION]: The skill is designed to package local repository context, including code snippets and architecture details, into a prompt for use with external services. This behavior is the primary intended purpose of the skill and requires manual user intervention (copy-pasting the output) to transmit data outside the local environment.
[PROMPT_INJECTION]: The skill reads from various local sources such as documentation, source files, and test results to build its prompt. This creates a surface for indirect prompt injection, as malicious instructions within those files could potentially influence the external model that receives the generated prompt. The skill does not currently specify the use of strict boundary delimiters or sanitization for this ingested content.

gpt-pro