The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the git branch --show-current command to retrieve the name of the active development branch. This is used solely to provide metadata for the context report and does not represent a security risk.
[PROMPT_INJECTION]: The skill processes untrusted content from local project files which could theoretically contain instructions designed to manipulate the AI. The skill mitigates this surface by using a separate subagent for summarization and explicitly constraining it to provide short summaries rather than raw data, which prevents external instructions from being directly interpreted by the primary agent context.

prime