high-end-visual-design
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a workflow that takes agent output and writes it to the local file system within the
llmwiki/directory. - Ingestion points: User-provided prompts and design requirements are ingested to generate UI code and reports (SKILL.md).
- Boundary markers: The skill lacks explicit instructions or delimiters to separate untrusted user input from the execution logic, creating a risk that malicious data could influence the files written to the wiki.
- Capability inventory: The skill utilizes file write operations to create draft documents and append entries to
llmwiki/wiki/index.mdandllmwiki/wiki/log.md(SKILL.md). - Sanitization: No procedures for validating, escaping, or sanitizing the content generated from user prompts before it is written to the file system are defined.
Audit Metadata