The Agent Skills Directory

[SAFE]: The skill instructions focus entirely on text generation and copywriting principles. It does not perform any network operations, execute code, or access sensitive system files.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by instructing the agent to read local context files such as .agents/product-marketing.md. However, since the skill is restricted to text generation and lacks dangerous capabilities like arbitrary command execution or network exfiltration, this surface is considered low risk and handled safely within the intended design.
Ingestion points: SKILL.md references .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md as context sources.
Boundary markers: Absent; the instructions simply direct the agent to read the context files without specific delimiters.
Capability inventory: The skill instructions are limited to generating and formatting text output and do not involve subprocess calls or network write operations.
Sanitization: None specified for the ingested content.

copywriting