form-vue
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates and composables for building Vue 3 forms that ingest untrusted user input, creating a potential surface for indirect prompt injection if the collected data is later processed by an LLM.
- Ingestion points: User input fields for credentials (email, password), usernames, and dynamic teammate information defined in
SKILL.mdandscripts/veevalidate-composables.ts. - Boundary markers: The templates do not include explicit instructions or delimiters to isolate user-provided data from agent instructions during downstream processing.
- Capability inventory: The code examples and composables include network capabilities via the
fetchAPI for asynchronous validation (e.g.,check-username) and form submission handlers. - Sanitization: The skill promotes best practices by using Zod schemas to validate the structure and format of input data before processing.
Audit Metadata