Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@TestOps MCP Server show me the latest test run results"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for TestOps 5.25. Lets AI assistants (Cursor, VS Code, Claude, OpenCode, Windsurf, Zed, Kilo Code) work directly with projects, launches, test plans, and test cases in TestOps.
Once connected, an AI assistant can:
Find projects, launches, and test plans
Create and stop test launches
Read, create, and edit test cases
Work with scenarios and custom fields
Search test cases using AQL
Related MCP server: TestOps MCP Server
npm install @syn7xx/testops-mcp-serverOr run without installing — via npx:
npx @syn7xx/testops-mcp-server --url https://your-testops.com --token your-tokenShort flags: -u (url), -t (token), -h (help).
You can also use environment variables:
export TESTOPS_URL=https://your-testops.com
export TESTOPS_TOKEN=your-token
npx @syn7xx/testops-mcp-serverCLI flags take priority over environment variables.
mcp.json in your project or home folder:
{
"mcpServers": {
"testops": {
"command": "npx",
"args": ["-y", "@syn7xx/testops-mcp-server", "--url", "https://your-testops.com", "--token", "your-token"]
}
}
}.vscode/mcp.json in your project (key is servers, not mcpServers):
{
"servers": {
"testops": {
"command": "npx",
"args": ["-y", "@syn7xx/testops-mcp-server", "--url", "https://your-testops.com", "--token", "your-token"]
}
}
}opencode.jsonc:
{
"mcp": {
"testops": {
"type": "local",
"command": ["npx", "-y", "@syn7xx/testops-mcp-server", "--url", "https://your-testops.com", "--token", "your-token"],
"enabled": true
}
}
}Or with environment variables:
{
"mcp": {
"testops": {
"type": "local",
"command": ["npx", "-y", "@syn7xx/testops-mcp-server"],
"environment": {
"TESTOPS_URL": "https://your-testops.com",
"TESTOPS_TOKEN": "your-token"
},
"enabled": true
}
}
}claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\):
{
"mcpServers": {
"testops": {
"command": "npx",
"args": ["@syn7xx/testops-mcp-server", "--url", "https://your-testops.com", "--token", "your-token"]
}
}
}Never commit real tokens. Use environment variables or local-only configs.
Tool | Description |
| List projects with pagination |
| Find project by exact or partial name match |
| Get project by ID |
| Get test case trees for a project |
Tool | Description |
| List launches for a project with pagination |
| Get a launch by ID |
| Create a test launch (with envVarValueSets support) |
| Stop a running launch |
| Launch summary: counts by status + progress |
| Flat paginated test results for a launch |
Tool | Description |
| List test plans for a project with pagination |
| Get test plan by ID |
| Test plan statistics: automated/manual counts and durations |
| Get test cases from a test plan |
| Create a launch from a test plan |
| Sync test plan with its source |
Tool | Description |
| Get test case by ID |
| Summary: flattened step strings, custom fields, tags |
| Full scenario JSON (steps + expected results) |
| Same as |
| Get custom field values for a test case |
| Update test case metadata (name, description, status, tags, etc.) |
| Update a single scenario step |
| Replace all steps in a scenario |
| Set custom field values |
| Search test cases using AQL query |
| List test cases in a project tree |
| Create a test case (with steps, tags, custom fields) |
Tool | Description |
| Custom field definitions for a project |
| Available values for a custom field |
If TestOps uses a self-signed certificate and Node.js fails TLS validation:
export NODE_TLS_REJECT_UNAUTHORIZED=0This disables certificate checks for all HTTPS requests in that process. Use only when necessary and in a trusted environment.
# Clone
git clone https://github.com/syn7xx/testops-mcp-server.git
cd testops-mcp-server
# Install dependencies
npm install
# Build
npm run build
# Run
npm start -- --url https://your-testops.com --token your-tokenFor working on the codebase, see CONTRIBUTING.md.
MIT
Create a test run (launch). Body requires name and projectId.
Inputs: name (required), projectId (required), tags, extra, +4 more.
Get a launch by ID
Inputs: launchId (required).
Launch run summary: counts by test status and progress (ready).
Inputs: launchId (required).
List launches for a project with pagination
Inputs: projectId (required), page, size, sort.
Flat paginated test results for a launch (rows: id, name, testCaseId, status, duration, etc.).
Inputs: launchId (required), page, size, sort, +2 more.
Close/stop a running launch (ends the run on TestOps).
Inputs: launchId (required).
Test tool to verify server connectivity
Inputs: no named parameters.
Find a project by exact or partial name match
Inputs: name (required).
Get a project by its ID
Inputs: id (required).
Get custom field definitions for a project with pagination. Returns { items, total, page, size } for chunked fetching.
Inputs: projectId (required), page, size.
Get available values for a custom field with pagination. Returns { items, total, page, size } so the model can fetch in chunks.
Inputs: projectId (required), customFieldId (required), page, size.
Get test case trees for a project. Use the returned tree IDs with testcase_list_in_tree.
Inputs: projectId (required).
List all projects with pagination
Inputs: page, size, sort.
Create a new test case in TestOps. Required: name and projectId. Optional: steps, customFields, tags, precondition, etc.
Inputs: name (required), projectId (required), tags, links, +13 more.
Get a test case by ID
Inputs: id (required).
Get custom field values for a test case
Inputs: id (required), projectId (required).
Aggregated card: name, description, tags, custom fields, and step texts as a simple list.
Inputs: id (required), projectId.
Normalized scenario JSON (scenarioSteps, step bodies, expected results).
Inputs: id (required).
Alias for testcase_get_scenario.
Inputs: id (required).
List test cases under a project tree with pagination. Use parentNodeId to page into a folder.
Inputs: treeId (required), projectId (required), page, size, +6 more.
Search test cases using AQL query with pagination
Inputs: rql (required), projectId (required), page, size, +2 more.
Replace all steps in a test case scenario
Inputs: id (required), steps (required).
Update test case metadata (name, description, status, tags, etc.). Does NOT change steps or custom fields — use testcase_set_scenario / testcase_update_custom_fields for those.
Inputs: id (required), name, tags, links, +10 more.
Update custom field values for a test case
Inputs: id (required), fields (required).
Update one scenario step by its step id.
Inputs: stepId (required), body, expectedResult.
Get a test plan by ID
Inputs: id (required).
Test plan statistics: automated/manual counts, durations. Useful before running to estimate effort.
Inputs: id (required).
Get test cases from a test plan with pagination
Inputs: id (required), page, size, sort.
List test plans for a project with pagination
Inputs: projectId (required), page, size, sort.
Create a launch from a test plan. Body requires launchName.
Inputs: id (required), launchName (required), tags, extra, +1 more.
Sync test plan with its source: refreshes the test case list in the plan to match the current source.
Inputs: id (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 31 tools
Each tool targets a distinct entity and action, with no overlapping purposes. Aliases like testcase_get_step for testcase_get_scenario are clearly documented as such.
All tools follow a consistent verb_noun pattern with underscores, e.g., project_list, testcase_create. Even multi-word actions like testcase_update_custom_fields maintain the pattern.
31 tools is on the higher side but well-scoped for a full-featured test management server. Each tool has a clear purpose and covers core workflows without being excessive.
The tool set lacks any delete operations for projects, test plans, launches, or test cases. This is a significant gap for CRUD completeness and will likely cause agent failures when deletion is needed.
Create and manage test cases, test runs and results in Testiny from your AI assistant.
Official MCP server for Qase — manage test cases, runs, suites, defects via AI tools.
MCP server for AI access to SmartBear tools, including BugSnag, Reflect, Swagger, PactFlow, QTM4J.
Connect AI assistants to your Tedix organization’s tasks, knowledge, apps, and results through MCP.
