code-quality-audit
Pass
Audited by Gen Agent Trust Hub on May 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches installation scripts for the Trivy and Gitleaks security scanners from their official GitHub repositories (aquasecurity/trivy and gitleaks/gitleaks). These are well-known technology organizations and the downloads are required for the skill's setup process.
- [REMOTE_CODE_EXECUTION]: Employs piped bash patterns (
curl | sh) in theinstall-tools.shscript to install official security binaries. This execution is limited to established security products from trusted sources. - [COMMAND_EXECUTION]: Orchestrates the execution of numerous local analysis tools (e.g., PHPUnit, PHPStan, PHPMD, ESLint, Jest, and Semgrep) through
Bashandddevenvironments. These operations are essential to its function as an auditing tool. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted codebase files to generate audit reports.
- Ingestion points: Analyzes user-provided source files in
web/modules/customandsrcdirectories across multiple scripts (e.g.,full-audit.sh,security-check.sh). - Boundary markers: Not present. The scripts pass raw file paths to tools without implementing explicit "ignore instructions" delimiters for the agent context.
- Capability inventory: The skill utilizes
Bashto perform file reads, execute shell commands, and conduct network operations for tool installation. - Sanitization: Content from external files is not sanitized or escaped before being processed by the underlying scanning engines.
Audit Metadata