sankey-diagram-creator

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill specifies dependencies on standard, well-known Python packages (plotly, pandas, and kaleido) through a requirements.txt file. These are sourced from the public PyPI registry, which is a standard but external source.
  • [PROMPT_INJECTION] (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) as it ingests untrusted data from external files.
  • Ingestion points: The from_csv method in SKILL.md allows the agent to load and process data from arbitrary file paths.
  • Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the documentation for handling CSV content.
  • Capability inventory: The skill includes methods to write files to the local filesystem (save_html, save_image), which could be misused if an attacker-controlled CSV influences the agent's behavior.
  • Sanitization: There is no documentation regarding the sanitization of labels or values extracted from CSV files before they are rendered into HTML or used in further agent reasoning.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:01 PM