marketing-claims-review
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes marketing copy provided by the user (either pasted directly or via file path), which could contain malicious instructions designed to influence the agent's behavior or bias its legal analysis.
- Ingestion points: Untrusted marketing text from landing pages, emails, ads, or file paths provided in the
argument-hint. - Boundary markers: Absent; the skill extracts claims directly from the provided text without specific delimiters or instructions to ignore embedded commands within the input.
- Capability inventory: The skill reads local plugin configuration files (
CLAUDE.md,matter.md), writes outputs to local directories, and has the capability to trigger external legal research tools or web searches. - Sanitization: No explicit sanitization or filtering of the input text is performed before it is classified and analyzed against the legal standards.
Audit Metadata