competitors
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a vulnerability surface by ingesting data from external sources and local configuration files, though it lacks dangerous capabilities to exploit this surface.
- Ingestion points: Reads local context files (
.agents/product-marketing.md,.claude/product-marketing.md,product-marketing-context.md) and external review sites (G2, Capterra, TrustRadius). - Boundary markers: Absent. The skill does not define specific delimiters to isolate external research data from the generation instructions.
- Capability inventory: The skill generates marketing copy in Markdown format. It does not have access to subprocess execution, arbitrary file writing, or network operations beyond simple reading.
- Sanitization: Absent. No explicit validation or filtering of external content is specified.
- [SAFE]: The skill follows best practices for content generation and research. It uses standard templates and refers to well-known, trusted industry services and tools (e.g., Notion, Salesforce, HubSpot). No patterns for code obfuscation, credential exfiltration, or persistence were detected.
Audit Metadata