cloudbase-declarative-deploy
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific configuration files (cloudbaserc) and environment files (.env), which creates a surface for indirect prompt injection where untrusted data could influence agent actions.
- Ingestion points: Reads cloudbaserc.json/yaml/js and .env files from the local workspace.
- Boundary markers: The skill requires a dry-run plan (deployPlan) and explicit confirmation (confirm=true) before applying changes.
- Capability inventory: Executes shell commands via deployBuild and performs cloud resource management via deployApply.
- Sanitization: Implements schema validation for configuration files against an official JSON schema.\n- [DYNAMIC_EXECUTION]: The skill supports JavaScript-based configuration files (cloudbaserc.js) and executes build commands defined in the configuration, which involves dynamic code execution as part of the deployment workflow.\n- [COMMAND_EXECUTION]: The skill executes user-defined commands specified in the configuration files through the deployBuild tool.
Audit Metadata