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