Skills
skills/theorcdev/8bitcn-ui/fumadocs-registry-integration/Gen Agent Trust Hub

fumadocs-registry-integration

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONNO_CODE
Full Analysis
  • Indirect Prompt Injection (LOW): The skill instructions allow an agent to ingest user-provided component metadata and write it into the project's registry.json file. * Ingestion points: User-provided component names, titles, and descriptions. * Boundary markers: Absent; no explicit delimiters or instructions to ignore embedded commands are specified. * Capability inventory: File-write operations to registry.json and creation of component files. * Sanitization: Absent; no validation or escaping logic is provided for the interpolated strings.
  • EXTERNAL_DOWNLOADS (LOW): The documentation includes instructions to execute remote packages using the pnpm dlx command. * Pattern: pnpm dlx shadcn@latest * Risk: Downloads and executes the latest version of the shadcn CLI from the npm registry. While standard for this workflow, it represents execution of remote code.
  • NO_CODE (SAFE): The skill itself contains only markdown instructions and JSON templates with no embedded scripts or binary files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:32 PM