khe-design-vocabulary
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates strictly as a lookup service using local markdown files. It lacks the capability to execute shell commands, perform network operations, or access sensitive system files. All functionality is confined to text processing within the provided design lexicon.- [INDIRECT_PROMPT_INJECTION]: The skill features a vulnerability surface where user-supplied descriptions are processed to determine the correct design term. However, the risk is mitigated by the following factors:\n
- Ingestion points: User-provided descriptions of UI effects (e.g., "the thing where...") in the
/khe-design-vocabularycommand.\n - Boundary markers: None explicitly defined in the prompt instructions.\n
- Capability inventory: The skill is limited to naming effects and providing glosses; it contains no subprocess calls, file-write capabilities, or network access tools.\n
- Sanitization: None specified, but the agent's output is restricted to a single line of text, effectively neutralizing the risk of complex instruction injection.
Audit Metadata