data-visualization
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external installation instructions and resources from the inference-sh organization on GitHub.\n- [COMMAND_EXECUTION]: The skill uses the belt CLI and npx to install and run visualization tools.\n- [DYNAMIC_EXECUTION]: The skill provides Python code templates that are executed dynamically via a remote executor (infsh/python-executor).\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to visualize external data, creating an attack surface for indirect prompt injection.\n
- Ingestion points: The --input parameter of belt app run commands in SKILL.md.\n
- Boundary markers: Absent.\n
- Capability inventory: Remote Python execution (belt) and file writing (plt.savefig).\n
- Sanitization: Detailed design rules and a 'Common Mistakes' section are provided to the agent to ensure accurate and non-misleading visualizations.
Audit Metadata