The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill points to external resources at github.com/mbzadev. This URL appears to typosquat the specified author name (mabzadev) by omitting the character 'a', which is a common technique used to redirect users to unauthorized or malicious repositories.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface by automatically reading local files for context without implementing sanitization or security boundaries. Ingestion points: The agent is instructed to read .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md. Boundary markers: No delimiters or safety instructions are used to isolate external file content from the system prompt. Capability inventory: Content from these files directly influences the agent's strategy and recommendations. Sanitization: No filtering or escaping is performed on the ingested data.

co-marketing