The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process untrusted external data provided by users, such as PDF files, Word documents, and resume screenshots, to extract content and layout information.
Ingestion points: User-supplied PDF, images, screenshots, and text files referenced in the SKILL.md workflow.
Boundary markers: The instructions do not specify the use of clear delimiters or warnings to prevent the agent from following instructions potentially embedded within the user-provided resume data.
Capability inventory: The skill has capabilities to write HTML files, execute local scripts via shell commands, and access network resources for icons.
Sanitization: There is no mention of specific sanitization or filtering logic applied to the extracted text before it is used in the prompt context.
[COMMAND_EXECUTION]: The skill instructs the agent to use shell commands for core functionality, specifically for template inlining and PDF generation.
Evidence: SKILL.md references the execution of scripts/inline-template.mjs and the command npm run export:pdf which utilizes scripts/export-resume-pdf.mjs.
[EXTERNAL_DOWNLOADS]: The skill fetches assets from external sources to style the generated resumes.
Evidence: The skill references fetching official SVG icons from LobeHub via @lobehub/icons or CDN resources as documented in references/default-resume-template.md.

make-resume