The Agent Skills Directory

[SAFE]: The skill contains no executable code or scripts. All files are markdown-based instructional or reference materials.
[DATA_EXFILTRATION]: The skill instructions include reading a local file named .claude/product-marketing-context.md to gather task-specific context. This is a standard practice for this type of agent and does not target sensitive user or system data.
[PROMPT_INJECTION]: While the skill ingests untrusted user input and external context files (potential surfaces for indirect prompt injection), it possesses no dangerous capabilities such as network access, file writing, or command execution that could be exploited.

copywriting