skills/danielmiessler/claude-code-plugins-plus/orchestrating-deployment-pipelines/Gen Agent Trust Hub
orchestrating-deployment-pipelines
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill gathers user requirements to generate production-ready code and configurations. This creates a surface where untrusted user input could lead to the generation of malicious deployment logic. \n
- Ingestion points: User requirement gathering processes described in SKILL.md. \n
- Boundary markers: No explicit delimiters for untrusted input are provided in the instructions. \n
- Capability inventory: Code generation for shell scripts (
deploy.sh,init_pipeline.sh) and CI/CD configurations (YAML/Kubernetes). \n - Sanitization: No specific sanitization or validation methods are described in the instructions. \n- [COMMAND_EXECUTION]: The skill structure includes scripts for automation tasks such as dependency installation, service management, and deployment execution, which are standard for infrastructure tools. \n- [NO_CODE]: The analyzed files consist of documentation and instruction sets but do not include the actual source code or implementation of the referenced shell or Python scripts.
Audit Metadata