document-pptx
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from sources like databases, APIs, and CSV files to generate presentation content. This creates an attack surface where untrusted data could potentially influence the agent's output. However, this is a core functional requirement of the skill and no malicious exploitation patterns were found.
- Ingestion points: Data ingestion examples are provided in assets/quarterly-review.md (SQLite, requests, and pandas).
- Boundary markers: Not explicitly defined in the provided code templates.
- Capability inventory: The skill includes file writing (pptx.save), network operations (requests.get), and subprocess execution (unoconv for PDF conversion).
- Sanitization: The provided examples do not include explicit input sanitization, which is expected for basic templates.
- [DYNAMIC_EXECUTION]: The skill provides Python and Node.js code templates in assets/pitch-deck.md and assets/quarterly-review.md. These templates are intended to be executed by the agent to automate document generation. This behavior is consistent with the skill's stated purpose of document automation and relies on standard, well-known libraries.
Audit Metadata