The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to present and potentially run npx commands for tool installation. Evidence: npx skills add pedroromeroluna/ai-first-product-skills in SKILL.md.
[EXTERNAL_DOWNLOADS]: The skill references an external package repository to install additional components. Evidence: References pedroromeroluna/ai-first-product-skills for the skills add command.
[INDIRECT_PROMPT_INJECTION]: The skill ingests a Discovery Brief and generates a Research Plan that is written to the local filesystem. This creates a surface where instructions embedded in the brief could influence the resulting file content. Ingestion points: Discovery Brief (referenced in SKILL.md). Boundary markers: None specified in the instructions. Capability inventory: Filesystem writes to research/<YYYY-MM-DD>-research-plan.md or the current directory (documented in SKILL.md). Sanitization: No specific filtering or sanitization steps are defined for the input data.

ux-research