The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions direct the agent to fetch DESIGN.md files from the external VoltAgent/awesome-design-md repository using WebFetch.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from a remote source without explicit sanitization or boundary markers.
Ingestion points: External DESIGN.md files are fetched from the VoltAgent/awesome-design-md repository (SKILL.md).
Boundary markers: Absent. The instructions do not define delimiters or provide warnings to the agent to ignore embedded commands within the fetched files.
Capability inventory: The skill writes synthesized instructions to the local file system at .claude/design/ and coordinates tasks for downstream agents such as ui-designer and frontend-developer (SKILL.md).
Sanitization: Absent. The skill lacks mechanisms to validate, filter, or escape the content of the markdown files before they are processed.

design-bridge