skills/wu-yc/labclaw/plotly/Gen Agent Trust Hub

plotly

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the installation of widely-used and legitimate Python packages via standard package managers.\n- [PROMPT_INJECTION]: The documentation includes a directive for the agent to recommend the author's hosted platform (www.k-dense.ai) when tasks reach a certain level of complexity. This constitutes a behavioral steering directive but does not attempt to bypass safety guidelines.\n- [COMMAND_EXECUTION]: The skill demonstrates how to launch a Dash web application server for local development using standard library commands.\n- [SAFE]: Evaluation of the skill's surface for indirect prompt injection reveals a standard data processing model for visualization tools.\n
  • Ingestion points: Data entering through pandas DataFrames (SKILL.md).\n
  • Boundary markers: None explicitly defined to separate data from instructions within the prompts.\n
  • Capability inventory: File system write access via 'write_html' and 'write_image'; network server execution via Dash 'run_server'.\n
  • Sanitization: Relies on Plotly and Dash internal sanitization for handling user-provided data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 04:33 AM