enhance-component

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local development commands including yarn link:create, yarn link:apps, and yarn dev. These commands are expected for the development workflow and operate on the local environment without fetching remote scripts for execution.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes documentation and example files that could contain malicious instructions.\n
  • Ingestion points: Documentation files (src/components/ui/[component]/docs/index.mdx) and component examples (src/components/ui/[component]/examples/) are ingested via the Read tool.\n
  • Boundary markers: There are no markers or instructions telling the agent to treat the content of documentation or examples strictly as data.\n
  • Capability inventory: The skill has shell access via yarn, the ability to search files via Grep, and permissions to read and modify source code.\n
  • Sanitization: No validation or sanitization is applied to the content of the read documentation or example files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 01:53 PM
Security Audit — agent-trust-hub — enhance-component