taiwan-traditional-chinese

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The maintenance script scripts/fetch_terms.py is designed to download terminology data from the official Wikibooks 'Mainland-Taiwan Computer Terminology Comparison Table' and the sysprog21/zhtw-mcp GitHub repository. These sources are reputable and well-known for providing linguistic data for the Taiwanese tech community. The downloads are used solely to update the local references/terms.csv file.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute the provided scripts/lint_zhtw.py Python script and standard shell commands like grep. These tools are used locally within the skill environment to perform mechanical checks on the agent's output, such as verifying punctuation usage, character density, and terminology compliance.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes functionality to ingest and process text data (drafts or existing project files) to identify stylistic violations. While this involves processing untrusted data, the risk is minimal because the analysis is performed using static regular expressions and predefined terminology lists rather than executing the content as instructions. The results are used solely as feedback for the agent to improve its stylistic accuracy.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 02:03 AM
Security Audit — agent-trust-hub — taiwan-traditional-chinese