scientific-critical-thinking
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute a local Python script
python scripts/generate_schematic.pyto generate diagrams. While this is part of the skill's intended functionality, it involves spawning a subprocess using a script file provided within the skill package. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because its core purpose is to process untrusted external content (scientific claims and research papers).
- Ingestion points: External scientific text, research methodology descriptions, and claims provided for analysis (SKILL.md).
- Boundary markers: Absent. The skill does not provide specific delimiters or instructions to ignore embedded commands within the analyzed material.
- Capability inventory: The agent has access to
Bash,Write, andEdittools, and is instructed to execute a local Python script. - Sanitization: Absent. There are no guidelines for escaping or validating the external content before it is processed by the agent.
- [EXTERNAL_DOWNLOADS]: The skill refers to an external website
www.k-dense.aiand another skillscientific-schematics. These are documented as vendor-owned resources (K-Dense Inc.) and do not involve automated or hidden remote code execution.
Audit Metadata