cloudbase-declarative-deploy

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes project configuration files such as cloudbaserc.json, cloudbaserc.yaml, cloudbaserc.js, and .env files. These files represent an attack surface where a compromised workspace could introduce malicious instructions. However, the skill mitigates this through mandatory user review of the deployment plan and target environment ID confirmation.\n
  • Ingestion points: cloudbaserc.json/yaml/yml/js and .env files (SKILL.md, references/config-schema.md).\n
  • Boundary markers: The skill requires an explicit confirm=true flag for deployment and includes a 'Deployment Gate' protocol that requires a mandatory user declaration (SKILL.md, references/plan-and-apply.md).\n
  • Capability inventory: Capabilities include local shell execution via deployBuild and cloud resource management via deployApply (SKILL.md).\n
  • Sanitization: The skill performs schema validation against an official JSON schema before execution (references/config-schema.md).\n- [COMMAND_EXECUTION]: The deployBuild tool executes shell commands defined in the cloudbaserc file (specifically buildCommand and installCommand). This is the intended primary purpose of the tool for building hosting artifacts locally. The risk is managed by the requirement that these commands be pre-defined in the project configuration.\n- [EXTERNAL_DOWNLOADS]: The skill references and validates project configuration against an official JSON schema hosted at https://static.cloudbase.net/cli/cloudbaserc.schema.json. This is a vendor-owned resource used for legitimate validation purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 03:14 PM