genius-thinking
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a highly structured brainstorming process based on recognized innovation frameworks and cognitive science principles.
- [COMMAND_EXECUTION]: The skill performs file-system writes to a dedicated workspace (
.hypercore/genius-thinking/[topic-slug]/). It creates several files includingflow.jsonfor state tracking and multiple markdown files for analysis and results. This behavior is consistent with the skill's stated purpose of providing an organized, multi-file analysis report. - [EXTERNAL_DOWNLOADS]: The documentation includes references to several well-known scientific and academic platforms such as Nature, ScienceDirect, arXiv, and Wikipedia. These links are provided for informational purposes regarding the frameworks used (e.g., Tree of Thoughts, DMN-ECN switching) and do not involve the downloading or execution of untrusted code.
- [PROMPT_INJECTION]: The skill ingests user-provided problem statements and constraints. While this untrusted data is used to generate prompts for sub-agents (using a Tree of Thoughts approach), the risk is mitigated by the structured nature of the analytical workflow and the intended use of the tool for creative ideation. There are no patterns suggesting attempts to bypass model safety filters.
Audit Metadata