The Agent Skills Directory

[SAFE]: The skill is purely instructional, providing strategic frameworks and templates for marketing planning without the use of executable scripts or remote code execution.- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by instructing the agent to read external project files for business context. While this is a standard context-awareness feature, it creates an ingestion point for untrusted data.
Ingestion points: .agents/product-marketing.md, .claude/product-marketing.md, and product-marketing-context.md in the local workspace.
Boundary markers: Absent. The skill relies on natural language instructions rather than technical delimiters to separate file content from the prompt.
Capability inventory: None. The skill does not utilize subprocess calls, file-write capabilities, or network operations, which significantly mitigates the risk of any injected instructions being executed.
Sanitization: Absent. The content of the ingested files is used as-is for context.- [DATA_EXPOSURE]: The skill reads project-specific marketing context files. This access is confined to the local workspace and is relevant to the skill's stated purpose. No access to sensitive system directories (e.g., .ssh, .aws) or credential files (e.g., .env) was found.

lead-magnets