agent-d3js-skill
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides code templates that ingest external data and render it using D3's
.html()method in tooltips (found insub-skills/tooltips.mdandassets/interactive-template.jsx). This represents a potential vulnerability surface where malicious data could execute scripts if not properly sanitized. - Ingestion points: Data objects passed to visualization functions in
sub-skills/tooltips.mdandassets/interactive-template.jsx. - Boundary markers: Absent in the code snippets.
- Capability inventory: DOM manipulation and event listeners via D3.js.
- Sanitization: Absent (uses
.html()instead of.text()). - [EXTERNAL_DOWNLOADS]: The skill references the official D3.js CDN in
sub-skills/1-set-up-d3js.md(https://d3js.org/d3.v7.min.js). This is a reference to a well-known, official service.
Audit Metadata