Zebrunner MCP Server

# MCP Zebrunner: Executive Summary **Author:** Maksim Sarychau **Current Version:** 5.15.0 **Last Updated:** December 2025 --- ## Overview I built **MCP Zebrunner** — a bridge between Zebrunner (our test management platform) and AI assistants like Claude, Cursor, ChatGPT, and IntelliJ IDEA. This allows QA teams to interact with their test data using natural language instead of navigating complex interfaces. --- ## The Problem I Solved Test teams spend significant time on repetitive, manual tasks: - **Switching between tools** — navigating dashboards, clicking through menus, copying data - **Searching for test cases** — finding specific tests across large projects with hundreds of suites - **Analyzing failures** — investigating why tests failed, reviewing logs, screenshots, and videos - **Generating reports** — compiling metrics for stakeholders and management reviews - **Identifying duplicates** — finding redundant test cases that waste execution time These tasks slow down testing cycles and reduce overall team productivity. --- ## What I Created A natural language interface for test management. Instead of clicking through dashboards, teams can now simply ask: | Instead of... | Now just ask... | |---------------|-----------------| | Navigating to test case details page | *"Get test case MCP-123 details"* | | Manually reviewing test failures | *"Why did test 5451420 fail?"* | | Building custom reports | *"Show me yesterday's test results"* | | Searching across suites | *"Find all login-related tests"* | | Analyzing launch results | *"Analyze launch 120906 failures"* | --- ## Key Capabilities I Implemented ### 🔍 Test Failure Analysis - Deep forensic analysis of failed tests - Automatic screenshot and video analysis with AI - Error classification and root cause suggestions - Comparison with last passed execution to identify what changed ### 📋 Test Case Management - Retrieve, search, and filter test cases - Validate test case quality against best practices - AI-powered improvement suggestions - Duplicate detection (both exact and semantic) ### 🚀 Launch & Execution Management - Comprehensive launch details and summaries - Platform-specific results (iOS, Android, Web) - Milestone and build tracking - Test run history and trends ### 📊 Reporting & Analytics - Bug analysis and top defects identification - Quality metrics and coverage analysis - Ready-to-paste Jira ticket generation - Management-ready markdown reports ### 🧪 Automation Support - Test code generation for multiple frameworks (Java, Python, JavaScript) - Automation readiness assessment - Coverage analysis between test cases and implementations - Framework detection and template matching ### 🎯 Feature-Based Aggregation (Latest) - Find all test cases related to any feature keyword - Group results by suite hierarchy - Generate automation tags for targeted test runs --- ## Who Benefits | Role | Key Benefits | |------|--------------| | **Manual QA Engineers** | Faster test case reviews, quality validation, improvement suggestions | | **Automation Engineers (SDETs)** | Code generation, automation readiness assessment, coverage analysis | | **Developers** | Quick understanding of test requirements, failure investigation | | **QA Managers** | Instant quality dashboards, duplicate detection, team metrics | | **Product Managers** | Project health overview, release readiness assessment, risk analysis | --- ## Current State ### Distribution - ✅ **npm Registry** — `npm install -g mcp-zebrunner` - ✅ **MCP Registry** — Listed in official Model Context Protocol registry - ✅ **Docker Hub** — `msarychau/mcp-zebrunner:5.15.0` - ✅ **GitHub** — Open source at github.com/maksimsarychau/mcp-zebrunner ### Scale - **40+ tools** covering all aspects of test management - **5 major integrations** — Claude Desktop, Cursor, ChatGPT Desktop (beta), IntelliJ IDEA, Docker - **3-tier intelligent rules system** for quality validation - **Multiple output formats** — JSON, Markdown, Jira-ready tickets ### Compatibility - Works with any Zebrunner instance - Supports all major AI assistants via MCP protocol - Cross-platform (macOS, Windows, Linux) --- ## Business Impact ### Time Savings | Task | Before | After | |------|--------|-------| | Test failure investigation | 15-30 min | 2-5 min | | Finding related test cases | 10-20 min | 30 sec | | Generating Jira tickets | 5-10 min | 1 min | | Weekly quality reports | 1-2 hours | 10 min | | Duplicate analysis | Manual/impossible | Automated | ### Quality Improvements - **Consistent validation** — All test cases checked against same quality standards - **Proactive duplicate detection** — Reduce redundant test execution - **AI-powered suggestions** — Continuous improvement recommendations - **Standardized reporting** — Same format across all teams and projects ### Accessibility - **No technical expertise required** — Anyone can query test data - **Natural language interface** — No need to learn query syntax - **Instant answers** — No waiting for someone to build reports --- ## Evolution Timeline | Version | Milestone | |---------|-----------| | v1.x | Basic test case retrieval | | v2.x | Suite hierarchy and batch operations | | v3.x | Test failure analysis and screenshots | | v4.x | Video analysis, Jira integration, intelligent rules | | v5.x | Docker support, MCP Registry, feature aggregation | --- ## Technical Foundation Built on the **Model Context Protocol (MCP)** — an open standard for connecting AI assistants to external tools and data sources. This ensures: - **Future-proof** — Works with any MCP-compatible AI assistant - **Secure** — Credentials stay local, no data sent to third parties - **Extensible** — Easy to add new capabilities - **Standard-compliant** — Following industry best practices --- ## What's Next Planned enhancements: - Enhanced semantic search across test cases - Predictive failure analysis - Test case generation from requirements - Integration with CI/CD pipelines - Enterprise SSO support (Okta, SAML) --- ## Quick Links - **Installation Guide:** [INSTALL-GUIDE.md](../INSTALL-GUIDE.md) - **npm Guide:** [MCP_NPM_INSTALLATION_GUIDE.md](../MCP_NPM_INSTALLATION_GUIDE.md) - **Docker Guide:** [DOCKER_USAGE.md](DOCKER_USAGE.md) - **Full Tools Catalog:** [TOOLS_CATALOG.md](../TOOLS_CATALOG.md) - **Change Log:** [change-logs.md](../change-logs.md) --- **Bottom line:** I created a tool that transforms how QA teams interact with their test data — making test management faster, smarter, and accessible to everyone through natural language AI assistants.