expo
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process external project files to automate tasks like code generation and design token synchronization, which presents an attack surface for indirect prompt injection.
- Ingestion points: The agent is directed to read
lib/api-spec/openapi.yamlinSKILL.mdfor running codegen andartifacts/<web-slug>/src/index.cssinreferences/design_and_aesthetics.mdto extract color and layout tokens. - Boundary markers: There are no explicit instructions for the agent to treat these file contents as untrusted data or to utilize boundary delimiters during processing.
- Capability inventory: The skill performs file writing, package installation, and execution of workspace-specific build tools (
pnpm run codegen). - Sanitization: No specific validation, filtering, or sanitization steps are defined for the content extracted from these project files.
Audit Metadata