sql-to-bi-builder

Warn

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the execution of several internal scripts (e.g., scripts/run_pipeline.py, scripts/setup_venv.sh) to process data and set up the environment.
  • [COMMAND_EXECUTION]: The skill dynamically generates a service bundle including a FastAPI backend and a frontend, along with shell scripts (start_backend.sh, start_frontend.sh) to run them. Executing generated code can be risky if the generation process is influenced by malicious input.
  • [PROMPT_INJECTION]: Potential for indirect prompt injection. The skill processes user-supplied markdown files (sql.md) containing SQL. Maliciously crafted SQL or markdown could attempt to manipulate the downstream code generation logic.
  • Ingestion points: Input markdown file (sql.md) parsed by scripts/parse_sql_md.py.
  • Boundary markers: None mentioned in the instructions to separate untrusted data from processing logic.
  • Capability inventory: File system access, script execution, and starting network services (FastAPI/frontend).
  • Sanitization: No mention of input sanitization or validation for the content of the SQL blocks.
  • [EXTERNAL_DOWNLOADS]: The skill requires installing external Python dependencies via requirements-dev.txt during the setup process.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 23, 2026, 04:58 AM
Security Audit — agent-trust-hub — sql-to-bi-builder