skills/gsmlg-dev/code-agent/denox/Gen Agent Trust Hub

denox

Warn

Audited by Gen Agent Trust Hub on Apr 27, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides numerous functions to execute arbitrary JavaScript or TypeScript code strings within an embedded V8 runtime.
  • [EXTERNAL_DOWNLOADS]: The skill supports fetching modules from external sources such as esm.sh, npm, and jsr registries during runtime evaluation or dependency management tasks.
  • [REMOTE_CODE_EXECUTION]: Code retrieved from remote URLs via dynamic imports is executed within the Deno runtime; while a sandbox option exists, executing remote code from untrusted sources is a significant security surface.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection if untrusted data is interpolated into the code strings passed to evaluation functions without proper sanitization.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 27, 2026, 06:32 AM
Security Audit — agent-trust-hub — denox