voltdb-development

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to verify that Docker is running (docker info) and to build and verify the generated project using Maven (mvn clean package, mvn verify). These operations are essential for a development tool that generates and tests code.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided descriptions of data models and custom tables to generate SQL DDL and Java stored procedures. This constitutes an indirect prompt injection surface where malicious input could influence code generation.
  • Ingestion points: User descriptions of tables, columns, and business logic collected in SKILL.md Step 5.
  • Boundary markers: None explicitly defined for separating user descriptions from the generation logic.
  • Capability inventory: The skill has the capability to write Java and SQL files to the local file system and execute shell commands via Maven and Docker.
  • Sanitization: The skill relies on natural language analysis to translate user requirements into code without explicit sanitization of the input strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 08:32 PM
Security Audit — agent-trust-hub — voltdb-development