ai-drama-prompt-factory
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill processes external text files (novels/stories) which serves as an indirect prompt injection surface. Maliciously crafted stories could attempt to influence the agent's output by embedding instructions within the narrative. However, the system's reliance on structured JSON schemas, explicit validation phases (Phase 4 Diagnosis), and clear transition protocols between phases effectively mitigates this risk.
- [COMMAND_EXECUTION]: The skill includes several Python scripts (
doc_indexer.py,doc_fetch.py,doc_search.py,doc_summarize.py) designed to handle large-scale text analysis. These scripts utilize standard libraries for file indexing, searching, and specific line-range retrieval. No unauthorized network operations, obfuscation, or privilege escalation mechanisms were found in these tools, which operate strictly on local file content provided for the workflow.
Audit Metadata