process-automated-builder

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. Untrusted data extracted from scientific articles and supplemental files (downloaded via scripts/origin/process_from_flow_download_si.py) is interpolated into multiple LLM prompts defined in tiangong_lca_spec/process_from_flow/prompts.py (e.g., EXCHANGES_PROMPT, EXCHANGE_VALUE_PROMPT). These templates do not utilize strict delimiters or specific instructions to ignore embedded malicious commands. This is notable as the LLM's structured output directly influences database actions via the Database_CRUD_Tool.
  • [EXTERNAL_DOWNLOADS]: The script scripts/origin/process_from_flow_download_si.py dynamically fetches literature and supporting data from doi.org and related academic publisher domains. While these are established sources, the skill automatically processes this external content.
  • [COMMAND_EXECUTION]: The workflow orchestrates multiple stages by spawning subprocesses using subprocess.run. For example, scripts/origin/process_from_flow_workflow.py executes a chain of internal scripts, and tiangong_lca_spec/tidas_validation/service.py invokes the tidas-validate tool. While these calls are directed at internal or known tools, they operate on data derived from the automated pipeline.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 01:38 AM