The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The file references/turso-setup.md contains instructions to download and execute an installation script from Turso's official domain via curl -sSfL https://get.tur.so/install.sh | bash.
[COMMAND_EXECUTION]: The skill generates an executable shell script at scripts/generate-image.sh (as detailed in references/image-generation.md) that uses curl for API interactions and python3 for processing JSON data.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface during its Discovery phase. * Ingestion points: Untrusted user input defining the site's niche, goals, and content in SKILL.md. * Boundary markers: No delimiters or specific instructions to ignore embedded commands are used for this input. * Capability inventory: The skill can execute shell commands, write files to the local filesystem, and perform network requests. * Sanitization: There is no evidence of validation or sanitization for the user-supplied strings before they are interpolated into project code.
[EXTERNAL_DOWNLOADS]: As described in references/image-generation.md, the skill fetches image data from Google's Gemini API at generativelanguage.googleapis.com to create custom assets.
[DYNAMIC_EXECUTION]: The skill's fundamental operation involves the dynamic generation of project files and scripts which are subsequently compiled or executed to build the user's website.

astro-builder