The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown-based instructional content, templates, and reference materials. There are no scripts, binaries, or automated tools included in the package.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for processing external data, specifically stakeholder interview notes and transcripts (e.g., in references/transcript-to-spec.md). While this constitutes a data ingestion surface, the skill provides structured methodologies for the agent to analyze and validate this data manually (classification of ACs as SAID, IMPLIED, etc.) and lacks any executable tools or shell access that could be targeted by a malicious payload embedded in the interview text.

product-discovery