research-backed-frontend-development
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill mandates that the agent perform online research, including viewing competitor products, screenshots, videos, and documentation. This creates a surface where an attacker could embed malicious instructions in those external resources to influence the agent's behavior.
- Ingestion points: Online searches for products, inspection of live websites, gallery images, help articles, and authoritative design documentation as described in the 'Research before designing' section of SKILL.md.
- Boundary markers: None identified. The skill does not provide instructions to delineate or treat external research data as untrusted.
- Capability inventory: The agent is empowered to browse the web, view media, and generate/propose code or design changes based on its findings.
- Sanitization: There are no instructions for the agent to sanitize, filter, or validate the content retrieved from external sources before incorporating it into its decision-making process.
Audit Metadata