scorecard-marketing

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of interactive funnels that ingest untrusted user input (quiz answers) and interpolate them into dynamic results pages, PDF reports, and automated emails. This pattern presents a surface for indirect prompt injection attacks.
  • Ingestion points: User-provided questionnaire responses described in SKILL.md and references/industry-examples.md.
  • Boundary markers: The templates for emails and result pages do not include explicit delimiters or instructions to ignore embedded commands within the user data.
  • Capability inventory: The skill documentation describes integrating results with CRM systems, generating PDF reports via Puppeteer, and sending automated emails in references/technical-implementation.md.
  • Sanitization: While the technical reference file provides examples for server-side validation and data sanitization, the prompts themselves rely on direct interpolation of user values.
  • [METADATA_POISONING]: There is a discrepancy between the author declared in the skill's YAML frontmatter (wondelai) and the provided author context (emmraan). This inconsistency in metadata attribution may lead to confusion regarding the source or ownership of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 11:36 AM
Security Audit — agent-trust-hub — scorecard-marketing