The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read local project context files (e.g., .agents/product-marketing.md or product-marketing-context.md) to inform its email designs. This creates a surface where untrusted content in those files could theoretically influence agent behavior.
Ingestion points: SKILL.md (Initial Assessment section) instructs the agent to read local marketing context files if they exist.
Boundary markers: There are no explicit delimiters or "ignore embedded instructions" warnings specified for the ingested file content.
Capability inventory: Based on the provided files, the skill's capabilities are limited to text generation and sequence planning. No subprocess calls, file-write operations, or network requests are defined in the instruction set.
Sanitization: No specific sanitization or filtering logic is provided for the ingested content.
[SAFE]: The skill focuses on content generation for email marketing. It correctly identifies its own scope by referring users to separate skills for cold outreach or in-app onboarding. No code, obfuscation, or dangerous execution patterns were detected.

emails