The Agent Skills Directory

[SAFE]: The skill is entirely instructional and provides a framework for planning product launches. It does not execute code, perform network requests (other than providing informational links), or access sensitive system resources.
[INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection because it reads context from external files such as .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md.
Ingestion points: The skill instructions in SKILL.md direct the agent to read the content of these marketing context files if they exist.
Boundary markers: The instructions do not specify any boundary markers or delimiters for the imported content.
Capability inventory: The skill does not possess high-risk capabilities; it primarily generates text-based marketing plans and does not have access to file-write, subprocess execution, or arbitrary network operations.
Sanitization: There is no specified sanitization for the content read from these files.
Note: Given the lack of dangerous capabilities, this surface is considered low-risk and does not elevate the overall verdict.

launch