configure-mock-server
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external JIRA tickets using the
jira_get_issuetool. This data, including endpoint details and schemas, is used to generate Java code and configuration files without explicit boundary markers or sanitization, potentially allowing malicious content in a ticket to influence agent behavior or the generated code. - Ingestion points: JIRA ticket description, custom fields, and comments.
- Boundary markers: Not specified in the instructions for handling JIRA data.
- Capability inventory: Writing Java files, modifying configuration YAMLs, and executing shell commands (
mvn,curl). - Sanitization: None identified for the extracted JIRA content.
- [DYNAMIC_EXECUTION]: The skill generates new Java classes (
DownstreamConfig.java,MockServiceProperties.java) based on templates and external input, and then instructs the agent to run the project usingmvn spring-boot:runfor verification. This constitutes execution of code generated at runtime. - [COMMAND_EXECUTION]: The skill executes shell commands (
curl) using URLs extracted from JIRA tickets to verify connectivity. This could be exploited if a malicious URL is provided in the JIRA ticket.
Audit Metadata