moai-foundation-quality

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The script scripts/quality-gate.sh dynamically generates a Python analysis tool using a heredoc and executes it at runtime via python3 -c. This pattern is used to bridge the shell environment with the Python-based quality engine.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external codebase data from user-specified directories (e.g., src/). This untrusted data is ingested for quality analysis. The skill possesses capabilities to write files (reports) and execute commands via the Bash tool, representing a potential attack surface for indirect injections. Evidence includes:
  • Ingestion points: Files within the path provided to analyze_codebase() or the quality-gate.sh script.
  • Boundary markers: None identified in the provided scripts or documentation.
  • Capability inventory: File writing (via QualityOrchestrator.generate_report), network operations (via Context7Client), and shell execution (via Bash tool and quality-gate.sh).
  • Sanitization: No specific content sanitization or instruction-ignoring delimiters are documented for the analysis phase.
  • [EXTERNAL_DOWNLOADS]: The skill and its integration templates (e.g., examples.md, github-actions-quality.yml) perform installations of external packages including moai-core-quality, moai-foundation-quality, eslint, and prettier from public registries. These are documented as vendor resources and standard development tools.
  • [COMMAND_EXECUTION]: The skill relies on the Bash tool to execute the quality-gate.sh script, which invokes system utilities such as find, jq, and python3 to facilitate the quality gate process.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 12:32 PM
Security Audit — agent-trust-hub — moai-foundation-quality