volt-kubernetes
Warn
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Instructions in SKILL.md (Step 1) direct the agent to run user-selected integration tests using Maven and Java to validate VoltDB components before deployment. This involves executing arbitrary code provided by the user in the agent's local environment.
- [DYNAMIC_EXECUTION]: The helmfile.tf implementation (referenced in SKILL.md Step 5) utilizes a null_resource with a local-exec provisioner to execute shell commands (helmfile apply). This pattern allows for the execution of arbitrary system commands.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided integration tests and configuration files. It lacks explicit sanitization or validation protocols for the content of these tests before the agent is instructed to execute them, creating a surface for malicious code execution through untrusted data ingestion.
Audit Metadata