wei-ran-chat
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates untrusted user input directly into its initial interaction, creating a potential surface for prompt injection attacks. • Ingestion points: The
$ARGUMENTSvariable inSKILL.mdis used to pass the user's first message directly to the AI persona. • Boundary markers: There are no explicit delimiters or boundary markers (such as XML tags or special tokens) used to isolate the user input from the persona's system instructions. • Capability inventory: The skill does not define or request access to high-privilege tools, network operations, or file system modifications, which mitigates the impact of potential injections. • Sanitization: No input sanitization or validation logic is implemented within the skill's instructions.
Audit Metadata