huawei-cloud-doris-sql-check

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests user-provided SQL via command-line arguments in the workflow python doris_sql_checker.py "<sql_text>". This pattern presents an indirect prompt injection surface where shell-metacharacters in the SQL could potentially trigger command injection in the executing shell environment.\n
  • Ingestion points: scripts/doris_sql_checker.py (receives input through sys.argv[1]).\n
  • Boundary markers: The workflow relies on shell-level double quotes which do not provide robust protection against malicious payloads.\n
  • Capability inventory: All scripts (doris_sql_checker.py, doris_sql_parser.py, doris_sql_tokenizer.py) perform local logic only, with no network or sensitive file system access.\n
  • Sanitization: No shell-specific sanitization is implemented for the command-line entry point.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:53 AM
Security Audit — agent-trust-hub — huawei-cloud-doris-sql-check