fasthtml
Fail
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: CRITICALPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: References documentation and repositories from the official FastHTML project (fastht.ml and github.com/AnswerDotAI). These are authoritative sources for the framework and consistent with the skill's purpose.
- [PROMPT_INJECTION]: The skill processes user-provided codebases to perform refactoring and review. It lacks explicit instructions to sanitize or delimit this content, creating a surface where malicious instructions within processed code could influence agent behavior. 1. Ingestion points: User-provided application code and project structure. 2. Boundary markers: None specified. 3. Capability inventory: Generation and modification of Python web application code. 4. Sanitization: No sanitization of ingested content is performed.
Recommendations
- CRITICAL: 1 infected file(s) detected - DO NOT USE
- Contains 6 malicious URL(s) - DO NOT USE
Audit Metadata