pmax-launch-kit

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an inherent vulnerability surface for indirect prompt injection because it ingests untrusted data from external websites via the WebFetch tool.
  • Ingestion points: SKILL.md Step 1 (reading content from the provided brand URL and key sub-pages).
  • Boundary markers: None explicitly defined to separate instructions from the fetched web content.
  • Capability inventory: The skill has access to Bash, Write, Read, WebSearch, and WebFetch tools.
  • Sanitization: While there is no explicit sanitization of fetched text, the skill instructions provide strong constraints (Step 3 and Step 8) requiring the agent to output structured data in specific tables with character count validation, which significantly mitigates the risk of an external site hijacking the agent's behavior for malicious purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 11:31 AM
Security Audit — agent-trust-hub — pmax-launch-kit