skills/tryopendata/skills/openchart/Gen Agent Trust Hub

openchart

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to fetch canonical type definitions and configuration files from remote sources, specifically unpkg.com and raw.githubusercontent.com. These references target the official @opendata-ai packages and the tryopendata/openchart repository, which are controlled by the skill's author and are used to ensure spec validity.
  • [DYNAMIC_EXECUTION]: The skill provides extensive references and code patterns for generating custom D3.js JavaScript and SVG code. This is an intended feature to support advanced infographics that go beyond declarative specifications, allowing the agent to generate and potentially facilitate the execution of complex visualization logic.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (such as SQL queries or provided data arrays) to generate charts and tables. This creates a surface where instructions embedded in the processed data could potentially influence the agent, although the structured VizSpec format helps mitigate accidental obedience.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 10:44 PM