The Agent Skills Directory

[COMMAND_EXECUTION]: The skill workflow includes standard shell commands (cat, ls, grep) to access brand guidelines, audience research, and image assets. These commands are restricted to local project directories (ai/memory-bank/, public/images/) and are consistent with the skill's purpose of analyzing project context.
[DATA_EXFILTRATION]: No network operations (such as curl or wget) or attempts to access sensitive system paths (such as .ssh or .env) were detected.
[PROMPT_INJECTION]: The instructions do not contain any patterns intended to bypass safety filters, override system constraints, or extract the agent's internal prompt. While the skill ingests content from external files in the memory bank, which is an indirect injection surface, the risk is mitigated by the lack of privileged capabilities or network access.
[SAFE]: The skill operates entirely within the local development environment using standard tools and follows its defined identity without introducing persistence or obfuscation.

design-visual-storyteller