novel-archiving
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests novel chapter content to generate summaries and update project state files such as characters and timelines. This creates a surface for indirect prompt injection if the processed text contains embedded instructions intended to manipulate the agent's behavior during the archiving steps.
- Ingestion points: The novel chapter text processed during the archiving flow (referenced as the chapter content).
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the novel text are provided.
- Capability inventory: The skill performs file write operations to project directories such as 'novel/' and '设定/' to maintain metadata.
- Sanitization: There is no evidence of sanitization or validation of the chapter content before it is used to update the archiving files.
Audit Metadata