cloudbase-declarative-deploy

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from local configuration files that influence tool behavior and command execution.
  • Ingestion points: Project configuration files (cloudbaserc.json, cloudbaserc.yaml, cloudbaserc.yml, cloudbaserc.js) and environment files (.env, .env.local, .env.<mode>) located in the project workspace.
  • Boundary markers: Implements mandatory confirmation flags (confirm=true, confirmDestructive=true) and references a mandatory Deployment Gate protocol before applying changes.
  • Capability inventory: Uses tools (deployBuild, deployApply, manageApps) capable of executing local shell commands, accessing filesystems for build artifacts, and performing network operations for cloud resource management.
  • Sanitization: Configuration is validated against the official CloudBase JSON schema (https://static.cloudbase.net/cli/cloudbaserc.schema.json) before processing.
  • [DYNAMIC_EXECUTION]: The skill facilitates the execution of local build and install commands defined within user-provided configuration files.
  • Evidence: The deployBuild tool and buildHint mechanism described in references/plan-and-apply.md resolve and execute commands (e.g., buildCommand) from the cloudbaserc configuration to prepare deployment artifacts locally.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 02:43 AM