The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the project environment, such as component source code and configuration files, to generate tests.
Ingestion points: The scripts functions/vr_setup_validator.py and functions/story_generator.py read content from local files including package.json, Storybook configuration, and frontend component files (React/Vue/Svelte).
Boundary markers: The instructions do not define delimiters or specific 'ignore' rules for the agent when reading and analyzing these external files.
Capability inventory: The skill leverages high-capability tools including Bash for script execution and Write for modifying project structure and CI/CD workflows.
Sanitization: There is no evidence of sanitization or filtering of the content read from component files before it is processed or interpolated into code generation templates.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute several bundled Python helper scripts. While these scripts perform legitimate project analysis, the use of a shell for execution represents a sensitive capability.

visual-regression