dx-kickstart-next
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the development of an application that ingests data from Contentstack APIs, which represents an indirect prompt injection surface if the agent processes CMS content at runtime.
- Ingestion points: The skill uses the Contentstack Delivery SDK (referenced in
lib/contentstack.ts) to fetch data from the Contentstack CMS. - Boundary markers: No explicit LLM delimiters are defined for data processing, though the instructions emphasize separating server-side rendering from client-side preview behavior.
- Capability inventory: The skill allows the agent to modify the filesystem (implementing changes) and execute shell commands for building and linting the project.
- Sanitization: The skill explicitly requires sanitizing all rich text HTML fields using
isomorphic-dompurifybefore rendering (documented inreferences/content-model.mdandreferences/contentstack.md). - [COMMAND_EXECUTION]: The instructions guide the agent to perform routine development operations, such as
npm install,npm run build,npm run lint, and project-specific CLI commands using the Contentstack CLI (csdx). These are appropriate for the skill's purpose as a developer tool. - [EXTERNAL_DOWNLOADS]: The skill involves installing dependencies from the npm registry and downloading stack seed configurations from Contentstack's official GitHub repositories. These are standard vendor-supported operations.
Audit Metadata