The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8). It processes untrusted user requirements to generate source code files.
Ingestion points: User-provided feature requirements and architectural plans in the prompt context.
Boundary markers: Absent; there are no specific instructions for the agent to ignore or delimit potentially malicious instructions within user inputs.
Capability inventory: The skill instructs the agent to create directories and write generated code artifacts to the filesystem (e.g., using mkdir -p .aix).
Sanitization: No explicit sanitization or validation of user-provided content is defined before it is used in code generation.
[COMMAND_EXECUTION]: The skill contains instructions for the agent to execute shell commands as part of its operational workflow.
Evidence: The 'AIX Pack Compliance' section in SKILL.md directs the agent to execute mkdir -p .aix to ensure the output directory exists before writing files.
Context: These commands are used for standard project structure management and do not involve privilege escalation or unauthorized access.
[EXTERNAL_DOWNLOADS]: The skill references several well-known and trusted JavaScript packages from the NPM registry.
Dependencies: Includes standard development tools such as vitest, pinia, vue, and eslint-plugin-jsdoc for code verification and testing.
Documentation: References additional guidance from the author's public documentation site (jeffallan.github.io).

vue-expert-js