shadcn-blocks

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md, the required runtime workflow performs “Search Blocks” via the ShadcnBlocks API using category/query and then “Preview & Select” candidate blocks, meaning outsider-authored free text can be ingested from API results before a specific item is necessarily preselected by the attacker.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill performs runtime network operations — e.g., querying https://api.shadcnblocks.com/v1/blocks?category=pricing&query=saas and installing blocks via npx using URLs like https://shadcnblocks.com/r/[block-id] (and references https://www.shadcnblocks.com) — which fetch remote code/content that is installed/executed and is required for the skill to function.