motion-figure
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests configuration data from markdown code fences to parameterize the rendering of figures. \n
- Ingestion points: Figure functions receive a
cfgobject containing JSON data parsed from markdown fences (e.g., inSKILL.mdandreferences/toolkit.md). \n - Boundary markers: Figures are delimited by standard markdown code block syntax. \n
- Capability inventory: The skill is restricted to DOM and SVG manipulation within a local host element. No file-system access, network operations, or high-privilege commands are present in the provided templates. \n
- Sanitization: Input is parsed as structured JSON and rendered using standard DOM methods, primarily
textContentand numeric attribute setting. \n- [DYNAMIC_EXECUTION]: The skill uses native browser animation APIs and runtime-compiled shaders for its visual components. \n - Evidence: The dither and mechanism templates utilize
requestAnimationFrame,setTimeout, and WebGL fragment shaders (GLSL) for rendering and interactivity. \n - These mechanisms are restricted to UI rendering and are standard for the skill's primary purpose of creating motion figures.
Audit Metadata