The Agent Skills Directory

[SAFE]: The skill operates as a purely instructional and template-based assistant for product management tasks. It defines strict classification logic and reporting structures without invoking dangerous shell commands, network operations, or unauthorized file access.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for the ingestion of user-provided feature lists and existing analysis documents (found in directories like product/ and research/). The risk of malicious instructions being processed as commands is mitigated by the skill's rigid adherence to a 5x5 evaluation table and structured markdown output. There is no evidence of autonomous action or sensitive data access linked to this ingestion surface.

kano-analysis