xd-extract-svg
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill uses
npx -y xd-extractto download and run the extraction utility directly from the NPM registry during execution. - [COMMAND_EXECUTION]: The agent is instructed to execute shell commands for the
xd-extractCLI (inventory, extract, validate, redact) and use system commands likeopenfor viewing preview files. - [EXTERNAL_DOWNLOADS]: The skill facilitates the retrieval of design manifests and scene-graph JSON payloads from Adobe XD (
xd.adobe.com) and other design viewer platforms using browser-based scraping methods. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external URLs and Adobe XD files, which serves as a potential attack surface.
- Ingestion points: User-provided Adobe XD share links and path to .xd files in SKILL.md.
- Boundary markers: No explicit boundary markers or security warnings for the design data are defined in the instructions.
- Capability inventory: Shell command execution via
npx, browser-based script evaluation (agent-browser eval), and local file system writes. - Sanitization: Relies on the
xd-extracttool's internal parsing and conversion logic to handle Adobe Graphic Content (AGC) safely.
Audit Metadata