The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-supplied tasks via the $ARGUMENTS placeholder. While the content is purely informational, the lack of boundary markers around these inputs presents a standard vulnerability surface when used with the skill's extensive toolset.
Ingestion points: User-provided tasks or questions are interpolated at the $ARGUMENTS placeholder in SKILL.md.
Boundary markers: Absent; there are no instructions or delimiters to isolate user input from the agent's core instructions.
Capability inventory: The skill has access to Bash, Write, Edit, Read, Grep, and Glob tools, enabling significant file system and shell interaction.
Sanitization: Absent; the skill does not specify any validation or filtering logic for the input arguments.
[EXTERNAL_DOWNLOADS]: The skill references several external resources relevant to the Even Hub G2 ecosystem, including an NPM package (even-toolkit) and GitHub repositories (even-g2-notes, even-toolkit). These resources are documented as community tools and official guidelines from the vendor's ecosystem and appear legitimate.

design-guidelines