The Agent Skills Directory

[SAFE]: The skill is designed as an interactive diagnostic tool for product management frameworks. It operates entirely within the chat context and does not invoke any external tools, scripts, or system commands. Analysis of the skill body and metadata confirms no malicious patterns such as prompt injection, obfuscation, or persistence mechanisms are present.
[INDIRECT_PROMPT_INJECTION]: The skill includes multiple user ingestion points, specifically the 'decision context' argument and responses to four adaptive questions. While it lacks boundary markers or sanitization for this input, the capability inventory for the skill is empty—there are no subprocess calls, network operations, or file-write functions to exploit. Consequently, the risk is negligible.
[DYNAMIC_CONTEXT_INJECTION]: No dynamic execution blocks (e.g., shell command interpolation) were detected in the skill instructions or metadata.

prioritization-advisor