The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill captures user input through the $ARGUMENTS variable and embeds it into a brief for a design tool. While this constitutes a data ingestion surface, the risk is minimal and inherent to the skill's function.
Ingestion points: User input from the $ARGUMENTS variable in SKILL.md is interpolated into the design brief.
Boundary markers: No explicit boundary markers or "ignore" instructions are used to separate user input from the rest of the brief.
Capability inventory: The agent is instructed to use its internal file-editing tools and the open-design MCP server's tools to generate and register files.
Sanitization: There is no explicit sanitization or input validation performed on the $ARGUMENTS string.

od-prototype-resume-modern