pptx-deck-builder
Pass
Audited by Gen Agent Trust Hub on Mar 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
pptxandexcalcommand-line tools to perform file operations, render diagrams, and build presentation decks. These operations involve executing subprocesses with parameters derived from workspace files. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes external data which could influence the agent's behavior.
- Ingestion points: Data is ingested from user-provided PowerPoint templates (
.pptx) via thepptx initcommand and from YAML specification files (deck.yaml) during the build process. - Boundary markers: The instructions do not define clear boundaries or 'ignore' directives to prevent the agent from following instructions potentially embedded within the template metadata or the YAML content.
- Capability inventory: The agent has the capability to execute shell commands, read local files, and write output files to the workspace.
- Sanitization: No sanitization or validation logic is specified for the text or metadata extracted from external templates before it is processed or incorporated into the final deck.
Audit Metadata