page-copy
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface related to the ingestion of untrusted external data, which is inherent to its browsing functionality.
- Ingestion points: The
SKILL.mdfile (Step 3 under 'Start with the business context') instructs the agent to fetch and read existing web pages via tools likeagent-browserto gather context for copywriting. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded in the fetched content.
- Capability inventory: The skill is limited to text and wireframe generation. It does not utilize subprocesses, file-write capabilities, or network-writing tools that could be abused via injection.
- Sanitization: No explicit sanitization or filtering of fetched data is mentioned.
- [SAFE]: No malicious patterns or security vulnerabilities were identified. The skill follows best practices by instructing the agent to use placeholders for unverified data and by maintaining a clear scope of operation limited to marketing assistance.
Audit Metadata