Ito is a runtime code review tool for pull requests (PRs). Its core idea is to verify whether the behavior of changes meets expectations by actually running the application before the code is merged, rather than relying only on static code analysis.
Purpose
Ito mainly solves the problem that code review can only read code diffs but cannot confirm runtime results. When a PR is opened, Ito reads the diff and PR description, automatically identifies the user flows affected by the changes, then builds and runs a real copy of the application in an isolated, disposable container environment, operates the application like a real user, and triggers real backend code execution, thereby discovering runtime errors that static review cannot catch, such as service authorization, concurrency handling, failure handling, data migration, identity verification, UI logic errors, or API integration failures.
Main capabilities
- Targeted test planning: After a PR is opened, Ito automatically reads the diff and PR description, focuses on the affected user flows, and autonomously builds a test plan, with no need to write test cases or maintain a test suite.
- Containerized test execution: Builds and runs a disposable real copy of the application for each PR, navigates the application like a real user, executes real backend code, and covers normal paths as well as edge cases and adversarial scenarios.
- Complete findings report within the PR: Each PR gets a complete test summary, including which flows passed, which failed, and Ito's investigation process. Each failure comes with video replay, relevant code lines, and reproduction steps, and provides a severity rating to help prioritize.
- Automatic rerun verification: After a fix is pushed, Ito automatically reruns to confirm the issue is resolved.
- Natural language instructions: Plain English test instructions can be added to prioritize important scenarios and skip unimportant ones.
- Security and compliance: Designed with security first, suitable for enterprises such as those in financial services, healthcare, and defense, providing SSO, audit logs, and dedicated support; tests run in isolated environments with end-to-end encryption and zero data retention; compliant with SOC 2 Type II, and independent audit reports are available upon request.
How to use
Setup is autonomous and requires no scripts. Just connect a GitHub repository, and Ito will map the application logic and begin testing within minutes. Each PR automatically gets a runtime report, including videos, screenshots, and failure details, displayed directly in the PR.
Applicable scenarios
- Teams use coding agents to produce a large number of PRs, and manual verification becomes a bottleneck.
- Want to let agents automatically merge code without adding a code review bottleneck.
- Need to catch runtime behavior regressions that static analysis cannot find.
- Want to reduce the time spent manually verifying AI-generated code and replace manual verification with automation and evidence.
Limitations and notes
- Official materials do not provide specific pricing information or charging model, so it is not available for now.
- Official materials do not provide specific operating system or deployment method restrictions for platform support, so it is not available for now.
- Official materials do not provide detailed information on free trials or version restrictions, so it is not available for now.
- The customer outcome data cited in the text (such as about 30% more features, 70% fewer production regressions, etc.) comes from official claims, and specific results may vary by team and usage.
