The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external source code and configuration files from the user's project environment to provide testing recommendations and refactoring.
Ingestion points: Quarkus Java source code, Maven pom.xml files, and application.properties configuration files.
Boundary markers: The skill includes specific "MANDATORY" and "SAFETY" instructions to run compilation checks and stop if errors are detected, providing a primitive validation of project state.
Capability inventory: The skill triggers the execution of build tool commands (mvn, mvnw) and requires a container runtime (Docker) for infrastructure provisioning.
Sanitization: No explicit sanitization or escaping of the user's code is performed before analysis, relying on the LLM's internal safety guardrails.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to manage the build lifecycle and verify test outcomes.
Evidence: Mandatory execution of ./mvnw compile or mvn compile before modifications and mvn clean verify after modifications.
Scope: These commands execute the build logic defined in the project's pom.xml, which could contain custom plugin configurations or execution hooks.

422-frameworks-quarkus-testing-integration-tests