Parasoft extends your AI coding agent with two complementary Java workflows: jtest-static-analysis for detecting and repairing static analysis violations, and jtest-unit-testing for generating and post-processing unit tests to improve coverage while maintaining a passing build.
Both skills are non-interactive, environment-variable driven, and designed for Maven or Gradle Java projects. They can be run independently, but they are typically most effective when used together: first fix priority violations, then generate or update tests around the changed code.
The skills integrate with the following coding agents using an installation script that automates setup:
- GitHub Copilot CLI
- OpenAI Codex CLI
Additionally, the skills can be integrated with any coding agent that supports MCP tools, skills, and sub-agents.
The jtest-static-analysis skill enables end-to-end automated static analysis and repair:
- Running Jtest static analysis against your Java project using any built-in or custom test configuration
- Collecting violations from the generated report, filtering them by severity, rule, or file scope, and ranking them in prioritized order
- Fixing violations automatically
- Verifying fixes by re-running Jtest static analysis incrementally after each change and comparing against the baseline report — ensuring that the same violation is resolved and no new violations are introduced
- Executing an optimized set of unit tests using TIA (Test Impact Analysis) to verify that the static analysis fixes do not break code functionality
- Committing fixes individually as separate Git commits
- Limiting scope to an entire project, package, or file, or to modified lines of code in a commit
The jtest-unit-testing skill enables end-to-end automated unit test generation:
- Running test creation against your Java project to generate new unit tests for the selected scope
- Verifying the created tests and fixing failures to ensure that they pass
- Controlling the number of fix attempts and maximum number of fixes to keep token usage under control
- Committing fixes individually as separate Git commits
- Limiting scope to an entire project, package, or file, or to modified lines of code in a commit
For details regarding installation and configuration of the skills, as well as their capabilities, usage examples, best practices, and CI/CD integration workflow, refer to the README document located at: https://github.com/parasoft/jtest-ai-agent-skills.
