The Agent Skills Directory

[NO_CODE] (SAFE): The skill consists entirely of Markdown files (SKILL.md and references/playbooks.md). There are no scripts, binaries, or configuration files that execute commands or logic on the host system.
[DATA_EXPOSURE] (INFO): The skill references reading a local file at .claude/product-marketing-context.md. While this involves file system access, it is a localized context-gathering action within the agent's environment and lacks any mechanism for data exfiltration.
[INDIRECT_PROMPT_INJECTION] (INFO): The skill defines a surface for processing user-supplied business context and keyword data. However, since the skill has no 'write' or 'execute' capabilities (no subprocesses, file modifications, or network calls), the risk associated with adversarial content in that data is negligible.

programmatic-seo