springboot-verification
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines several phases that execute shell commands to manage the Spring Boot lifecycle. This includes building the project (
mvn clean verify,./gradlew assemble), running static analysis plugins (spotbugs,pmd,checkstyle), and executing unit and integration tests. These commands are standard for the skill's stated purpose of project verification. - [INDIRECT_PROMPT_INJECTION]: The skill processes external project data which creates a surface for indirect prompt injection if the source code or configuration files contain malicious instructions meant to influence the agent's behavior.
- Ingestion points: The skill reads Java source code, configuration files (e.g., pom.xml, application.yml), and git history during the build and security scan phases.
- Boundary markers: The skill does not specify delimiters or provide instructions to the agent to treat the ingested source code as untrusted data.
- Capability inventory: The skill uses shell tools including Maven, Gradle, and grep to process the project files.
- Sanitization: There is no evidence of sanitization or filtering of the file content before it is processed or summarized in the verification report.
Audit Metadata