The Agent Skills Directory

[SAFE]: The skill consists exclusively of Markdown-based instructions and playbooks for marketing tasks. It does not perform any sensitive file access, network requests, or command execution.
[INDIRECT_PROMPT_INJECTION]: The instructions suggest reading local context files, such as .agents/product-marketing-context.md, to gather product information. This represents a data ingestion surface; however, it is determined to be safe because the skill lacks any capabilities (no file writing, no network access, no dynamic code execution) that could be used to exploit or exfiltrate the ingested data. The files referenced are standard local context stores for AI agents.

programmatic-seo