starter
Fail
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: CRITICALEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill mentions GitHub Pages and Vercel for project deployment. The URL 'https://username.github.io/repo-name' is an instructional placeholder for a well-known service and does not represent a malicious download or execution risk.\n- [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection.\n
- Ingestion points: The skill utilizes WebFetch and WebSearch to retrieve external content from the internet.\n
- Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted web content from agent prompts.\n
- Capability inventory: The skill allows the agent to Write and Edit local files, which could be exploited by malicious instructions embedded in external web content.\n
- Sanitization: There is no explicit sanitization or validation logic provided for data retrieved from external sources.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata