The Agent Skills Directory

[SAFE]: The skill provides structured instructions for content strategy tasks such as identifying pillars and researching keywords. It does not include any scripts or commands that pose a security risk.
[DATA_EXPOSURE]: The skill mentions reading a local context file, .claude/product-marketing-context.md. This is an established practice for context-aware agents and is not a sensitive file path.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data like transcripts and search results. While this is an ingestion point for external instructions, the skill has no access to dangerous capabilities (e.g., file writes, remote execution) that would make such an injection exploitable.

content-strategy