zephyr-mcp
Enables reading test cases and test steps from Zephyr Scale using the Zephyr Scale API.
README
zephyr-test-zuellig-agentic
Agentic test generation for eZRX+ — converts Zephyr Scale test cases into Playwright integration tests using Claude Code or Codex.
What this repo does
- Runs a Zephyr Scale MCP server to fetch test cases and steps.
- Runs a code-review-graph MCP server to index the frontend codebase for token-efficient context loading.
- Provides a
/write-integration-testcommand that Claude Code reads from.claude/commands/— no global config needed.
All generated tests are written into datasource/frontend/ (a local clone of
the frontend repo). Nothing is installed or modified in the real frontend repo.
Requirements
- Node.js 20.6 or later
- Python 3.10 or later with
code-review-graphinstalled:pip install code-review-graph - Claude Code CLI or Codex CLI
- Git Bash or WSL on Windows
Setup (first time)
1. Clone and enter the repo
git clone <repository-url>
cd zephyr-test-zuellig-agentic
2. Create .env
Get a Zephyr Scale API token:
- Sign in to Jira.
- Open profile menu → Zephyr Scale API Access Tokens.
- Create a token and copy it.

Then:
cp .env.example .env
Fill in the values:
ZEPHYR_API_TOKEN=your-zephyr-api-token
FRONTEND_REPO_URL=https://github.com/your-org/frontend.git
.env is gitignored — do not commit your token.
3. Run setup
./setup.sh
This will:
- Clone
FRONTEND_REPO_URLintodatasource/frontend/ - Build the code-review-graph index for the frontend
- Build the Zephyr MCP server from source
- Generate
.mcp.jsonfor Claude Code project-level MCP
4. Open in Claude Code
Open the zephyr-test-zuellig-agentic/ folder in Claude Code (not datasource/frontend/).
Claude Code auto-loads .mcp.json — both MCP servers start automatically.
Verify:
/mcp
Both zephyr and code-review-graph should show as connected.
Daily usage
Write a test from a Zephyr TC
/write-integration-test EZRX-T10000
Claude will:
- Fetch the TC name, objective, and steps from Zephyr
- Search the codebase graph for relevant page objects and similar specs
- Generate the spec file (and any new page object methods / locators needed)
- Run
tsc --noEmitandplaywright test --listto verify
Generated files land in datasource/frontend/integration_test/. Commit and push from there.
Update the graph after local changes
When you edit files in datasource/frontend/ and want Claude to see the changes:
./update-graph.sh
Pull latest frontend + update graph
git -C datasource/frontend pull --ff-only
./update-graph.sh
Troubleshooting
.mcp.json missing — run ./setup.sh to regenerate it.
datasource/frontend missing — run ./setup.sh to clone the frontend repo.
MCP not connected — restart Claude Code after setup.sh completes, then check /mcp.
ZEPHYR_API_TOKEN error — make sure .env exists and ZEPHYR_API_TOKEN is set.
code-review-graph not found — run pip install code-review-graph.
Development
npm run typecheck
npm run build
Zephyr MCP source is in src/; compiled output goes to build/.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.