The Agent Skills Directory

[SAFE]: The skill body consists entirely of natural language instructions and templates for generating test cases. It does not invoke any shell commands, scripts, or external tools.- [INDIRECT_PROMPT_INJECTION]: While the skill ingests external data through the $USER_STORY and $CONTEXT arguments, it lacks any functional capabilities (such as file writing, network access, or command execution) that could be exploited by malicious content within those inputs. The ingestion surface is limited to text generation.- [SAFE]: No obfuscation, hidden URLs, or malicious metadata patterns were identified during the analysis of the skill file.

test-scenarios