resend-design-skills
Pass
Audited by Gen Agent Trust Hub on May 6, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The
design-auditskill processes untrusted data by reading project source code fromsrc/app/(dashboard)*andsrc/ui/to identify design system violations. This surface is vulnerable to indirect prompt injection where malicious instructions embedded in code comments could influence the audit report or the content of tickets filed in Linear. - Ingestion points: Reads files from the local repository as specified in
design-audit/SKILL.md. - Boundary markers: The instructions do not define delimiters or specific "ignore embedded instructions" warnings for the ingested code.
- Capability inventory: The skill uses Linear MCP tools (
linear:create_issue,linear:create_comment) to exfiltrate audit findings to a project management board. - Sanitization: There is no explicit requirement for the agent to sanitize or escape project content before including it in Linear ticket bodies.
Audit Metadata