Where are the test results and screenshots stored?

All data, including test definitions, user feedback, and uploaded images, are stored locally on your machine in the ~/.e2e-tester/ directory.

How does the human-AI separation work in this skill?

The user handles subjective tasks like UI navigation and visual verification, while Claude handles objective technical tasks like database lookups, log grepping, and API status checks.

Can Claude automatically fix issues found during testing?

Yes, once the automated and manual results are analyzed, Claude can enter plan mode to investigate and fix any detected failures in the code.

Is this skill suitable for testing production environments?

It is primarily designed for development and staging environments where Claude has the necessary permissions to run CLI commands and inspect backend systems.

Do I need Docker to use the E2E Test Orchestrator?

Yes, Docker is required to host the local web application where users track their manual testing steps and upload screenshots.

E2E Test Orchestrator

Name: E2E Test Orchestrator
Author: SaharCarmel

bySaharCarmel

•

Security & Testing

Coordinates end-to-end testing by delegating UI actions to humans and backend verifications to Claude.

The E2E Test Orchestrator bridges the gap between manual visual verification and automated backend validation. It generates structured test checklists where users perform UI-based tasks like clicking buttons or taking screenshots, while Claude simultaneously executes complex backend checks such as database queries, API calls, and log analysis. This hybrid workflow ensures comprehensive feature coverage and faster debugging by identifying exactly where a failure occurs—whether in the user interface or the underlying system architecture.

Key Features

01Human-AI Hybrid Workflow: Separates visual UI actions for users and technical verifications for Claude.

022 GitHub stars

03Automated Backend Validation: Claude executes CLI commands, DB queries, and API calls to verify data integrity.

04Unified Results Reporting: Aggregates manual observations and automated results into a single diagnostic summary.

05Contextual Test Generation: Automatically creates feature-specific test suites based on implemented code.

06Docker-Powered Web Interface: Launches a local web app for users to track steps and upload evidence.

Use Cases

01Performing user acceptance testing (UAT) with deep backend auditing.

02Regression testing to ensure UI changes haven't broken database or API logic.

03Verifying complex feature implementations before merging pull requests.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add saharcarmel/sahar-claude-code-marketplace e2e-tester

For use in Claude.ai and ChatGPT

Download Skill