conceptual-framework-builder
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied academic content (research questions, literature, and theoretical concepts) to generate a final conceptual framework document. This creates an inherent attack surface where malicious instructions embedded in the user's research data could potentially influence the agent's behavior during the integration or file-saving process. However, this is a functional requirement for the skill's stated purpose of document generation.
- [COMMAND_EXECUTION]: The skill includes instructions to save a final integrated markdown document to the user's desktop or a specified location. This is a legitimate file-write operation consistent with the skill's utility as a framework builder.
- [REMOTE_CODE_EXECUTION]: The skill mentions invoking an external capability (
/baoyu-infographic) to generate a visual concept map. This represents a standard inter-skill interaction or tool call within the agent environment to provide a visualization feature for the academic framework.
Audit Metadata