The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill performs standard image processing and file management tasks using common tools like ImageMagick.
[COMMAND_EXECUTION]: The skill uses shell commands (magick, cp, mkdir) which are correctly scoped to the project environment and are used to fulfill the primary purpose of the skill. Shell variables are handled with proper quoting.
[DATA_EXPOSURE]: The skill reads local project metadata files such as package.json and site.webmanifest to extract configuration details like the application name. This is standard behavior for project-integrated tools and does not involve sensitive user credentials.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files to determine its configuration.
Ingestion points: Reads package.json, site.webmanifest, and directory names to determine the application name.
Boundary markers: Not explicitly defined, but the extracted data is used only for templating (e.g., in manifest strings or HTML tags).
Capability inventory: Includes image processing (magick), file copying (cp), and file creation/modification (manifests and layout files).
Sanitization: Validates the source image file extension before processing.

favicon