jobkorea-talent-search
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions suggest using
npxto fetch updated instructions and CLI tools from the@nomadamas/k-skillpackage hosted on the npm registry, a well-known package service. - [COMMAND_EXECUTION]: The documentation provides shell commands for the user to execute for updating the skill environment and running fallback search scripts.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests talent profile data from the JobKorea website. To mitigate risks from processing external data, the script implements text sanitization by removing
<script>and<style>tags via regular expressions before the content is parsed or presented to the agent. - [DYNAMIC_EXECUTION]: The test suite (
scripts/test_jobkorea_talent_search.py) utilizesimportlib.utilto dynamically load the search module for verification purposes using static local paths.
Audit Metadata