The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external files or folders to generate web content, which presents a surface for indirect prompt injection where data content could potentially influence agent behavior during processing.
Ingestion points: Processes data from user-provided file paths or folders as specified in the workflow.
Boundary markers: No specific delimiters or "ignore instructions" wrappers are defined for the source materials.
Capability inventory: The skill possesses the ability to execute local scripts (JavaScript and Python) for rendering and verification, and writes files to the local filesystem.
Sanitization: No explicit sanitization or validation of the input data is mentioned in the instructions.
[COMMAND_EXECUTION]: The workflow involves the execution of local utility scripts to verify and package the deliverables.
Evidence: References execution of capture_page.mjs, export_pdf.mjs, visual_lint.mjs, image_bounds.py, and make_contact_sheet.py located in the .shared\visual-runtime directory.

web-artifacts-builder-codex