The Agent Skills Directory

[SAFE]: The skill is purely informational, consisting of Markdown-based frameworks and templates. It does not include scripts, binaries, or configuration that would allow for command execution or network access.
[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for processing external input (stakeholder interview transcripts). While this is a data ingestion surface, the lack of automated tool execution or dangerous capabilities minimizes the risk of indirect prompt injection. The documentation explicitly recommends human review gates and stakeholder validation loops.
[NO_CODE]: No executable code, scripts, or system-level configuration files are provided. The content is strictly limited to markdown guidelines, templates, and evaluation data.

product-discovery