novelist-style
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from multiple external project files and user-provided reference texts to generate a style guide, which is a potential surface for indirect prompt injection.
- Ingestion points: The agent reads content from
00-概述.md,01-世界观设定.md,02-人物设定.md, and arbitrary reference text provided by the user. - Boundary markers: The instructions do not include specific delimiters or instructions to the agent to disregard potential commands embedded within the input data.
- Capability inventory: The skill has the capability to write new files to the filesystem (e.g.,
08-文字风格.md). - Sanitization: There is no evidence of input validation, filtering, or escaping to sanitize the ingested content before it is processed by the agent.
Audit Metadata