The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions are task-oriented and do not contain patterns intended to bypass agent guardrails or override core instructions.
[DATA_EXFILTRATION]: No unauthorized network operations or sensitive file access patterns were found. The skill explicitly includes a guardrail advising against the storage of secrets or credentials within the generated project files.
[EXTERNAL_DOWNLOADS]: The skill refers to the Azure Retail Prices API for retrieving pricing data. This is a well-known and legitimate service for the skill's stated purpose of performing cost comparisons.
[COMMAND_EXECUTION]: Operations are restricted to creating directories and markdown templates. Potential risks from user-provided folder names are mitigated by instructions to convert topics into English kebab-case slugs.
[INDIRECT_PROMPT_INJECTION]: The skill processes user input to populate documentation templates. It follows a structured approach for generating content, which limits the attack surface for indirect injection by using predefined schemas for validation plans and cost comparisons.

project-workspace