drama-evaluator
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by ingesting and analyzing user-provided story texts, although the lack of capabilities renders this surface non-exploitable.
- Ingestion points: Instructions in
SKILL.md(e.g., "深入阅读和理解故事文本") explicitly state the skill should process external story content supplied by the user. - Boundary markers: The skill does not provide delimiters or negative constraints to prevent the model from obeying instructions embedded within the user's story text.
- Capability inventory: There are no scripts, environment configurations, or
allowed-toolslisted, meaning the skill has no ability to write files, execute shell commands, or perform network requests. - Sanitization: No sanitization or filtering logic is prescribed for the input story text before the evaluation process begins.
Audit Metadata