sn-ppt-story
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from sources like 'raw_documents.json' and 'research_report' without explicit instruction boundary markers. This data is synthesized into an 'outline.md' file that governs downstream presentation generation, creating a surface for indirect prompt injection attacks.
- Ingestion points: '<DECK_DIR>/info_pack.raw_documents', '<DECK_DIR>/info_pack.research_report', and other user-specified source files.
- Boundary markers: The instructions do not specify delimiters or 'ignore instructions' warnings to prevent the agent from obeying commands embedded in the processed documents.
- Capability inventory: File system write access to the project directory to update 'outline.md' and 'task_pack.json'.
- Sanitization: No sanitization, escaping, or schema validation is mentioned for the ingested document content.
Audit Metadata