upload-element-previews

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using bun run upload-element-preview. This is used to trigger a local script within the packages/docs directory to process and upload assets.
  • [DATA_EXFILTRATION]: The skill is designed to upload local preview assets from .element-previews or static/elements to an external domain remotion.media. This domain is a known resource associated with the vendor 'remotion-dev' and is used for hosting project-specific media.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied or project-derived data (element slugs and categories) which are then interpolated directly into a shell command line.
  • Ingestion points: The 'Element slug' identified in step 1 of the instructions.
  • Boundary markers: No explicit boundary markers or 'ignore' instructions are provided for the input slug.
  • Capability inventory: The skill possesses the capability to execute shell commands (bun run), delete local files (rm via instruction 5), and perform Git operations (commit and push).
  • Sanitization: The instructions do not specify sanitization or validation logic for the slug input before it is used in the bun command.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 06:47 PM