The Agent Skills Directory

[SAFE]: The skill follows best practices for technical documentation, including verification checklists and citation requirements. It focuses on analyzing project-specific code and official documentation without attempting to access sensitive system files or perform network operations outside its defined scope.
[PROMPT_INJECTION]: The skill processes user-supplied topics and context through the $ARGUMENTS variable, which creates an indirect prompt injection surface. A malicious user could provide input designed to influence the agent's research or output generation process.
Ingestion points: $ARGUMENTS in SKILL.md used to define the blog topic and context.
Boundary markers: Absent; the input is used directly as the basis for the generation process.
Capability inventory: Read, Write, Glob, Grep, and Bash tools are available to the agent as defined in SKILL.md to facilitate code analysis.
Sanitization: Absent; the agent relies on the validation steps in SKILL.md and the GUIDELINES.md checklist to ensure accuracy and integrity.

tech-blog