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., "@Test Reporter MCP Server Why is checkout.spec.js flaky?"
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 (Model Context Protocol) server for Test Ledger that enables Claude Code to analyze flaky tests, find failure patterns, suggest fixes and more.
No installation required! Just add the configuration to Claude Code.
Related MCP server: Tesults MCP
Log into testledger.dev and go to Settings → API Keys to generate a key.
Add this to your Claude Code MCP config:
Location: ~/.claude.json (global) or .mcp.json (project)
{
"mcpServers": {
"test-reporter": {
"command": "npx",
"args": ["-y", "@testledger/mcp@latest"],
"env": {
"TEST_LEDGER_API_KEY": "your-api-key-here"
}
}
}
}Tip: Using
@latestensures you always get the newest version with bug fixes.
That's it! Claude Code now has access to your test results.
Once configured, you can ask Claude Code things like:
"Why is checkout.spec.js flaky?"
"What tests have been failing the most this week?"
"Show me recent test failures"
"Are there any tests that always fail together?"
For the best experience, add the fix-flaky-test slash command to your project:
mkdir -p .claude/commands
curl -o .claude/commands/fix-flaky-test.md https://raw.githubusercontent.com/your-company/test-reporter-mcp/main/commands/fix-flaky-test.mdThen use it:
/fix-flaky-test
Test: LoginPage.should allow user to login with valid credentials
Error: element ("#submit-btn") still not clickable after 3000ms
at login.spec.js:42:24The MCP server provides these tools to Claude:
Tool | Description |
| Pass/fail/flaky statistics for a test |
| Time-of-day, browser, and version patterns |
| Tests that fail together (shared setup issues) |
| Project-wide flaky test leaderboard |
| Recent failures for quick triage |
| Failure rate over time |
Environment Variable | Required | Description |
| Yes | Your API key from the dashboard |
| No | Custom API URL (default: |
| No | Default project ID to use for queries |
{
"mcpServers": {
"test-reporter": {
"command": "npx",
"args": ["-y", "@testledger/mcp"],
"env": {
"TEST_LEDGER_API_KEY": "tr_live_abc123",
"TEST_LEDGER_PROJECT_ID": "42"
}
}
}
}Restart Claude Code after updating config
Check for JSON syntax errors in your config file
Verify your API key is valid
Your API key is invalid or expired. Generate a new one from the dashboard.
Your API key doesn't have access to the requested project. Check project permissions.
Documentation: testledger.dev
Issues: GitHub Issues
Get tests that are failing consecutively (broken tests, not flaky). Returns tests where the last 2+ runs have failed, with timing info (last_passed_date, first_failed_date) useful for identifying which merge broke them.
Inputs: days, limit, version, project_id, +1 more.
Analyze when and how tests fail to identify patterns. Returns failure rates by hour, day of week, version, browser/site, and duration analysis.
Inputs: spec_file (required), days, project_id, test_title.
Get screenshots from recent test failures. Returns presigned S3 URLs that can be viewed with the Read tool to see exactly what the UI looked like when the test failed.
Inputs: spec_file (required), days, limit, project_id, +1 more.
Get flaky specs from pre-computed materialized view. Faster than get_flaky_tests as it uses cached data refreshed hourly. Returns spec-level flakiness (not individual test level).
Inputs: limit, project_id, min_total_runs, min_flaky_count, +1 more.
Get a list of flaky tests (tests that fail then pass on retry) across the project, sorted by flakiness rate. Note: This scans all tests - use smaller 'days' values for faster results.
Inputs: days, limit, project_id, min_flaky_rate.
Get the most recent test failures for quick triage. Useful for seeing what's currently broken. For faster results, provide a spec_file filter.
Inputs: hours, limit, spec_file, project_id.
Get historical pass/fail/flaky statistics for a specific test. Use this to understand how often a test fails and its overall reliability. Returns health_status (healthy/flaky/broken/disabled/insufficient_data) from the test_health view for AI pre-filtering decisions.
Inputs: spec_file (required), days, project_id, test_title.
Get trend data for a test over time, useful for seeing if a test is getting more or less reliable.
Inputs: spec_file (required), days, project_id, test_title, +1 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 9 tools
Most tools target distinct analytical questions (patterns, correlation, screenshots, consecutive failures). The main overlap is get_flaky_tests vs get_flaky_specs (test-level vs spec-level) and some tension between get_test_history and get_test_trend, but the descriptions explicitly clarify the differences. Boundaries are mostly clear.
Every tool follows the consistent get_<entity> snake_case pattern (get_failure_patterns, get_test_history, get_flaky_tests, etc.). Naming is fully predictable with no stylistic deviations.
Nine tools is well within a sensible range for a test analytics domain and each tool maps to a distinct analytical query. No bloat or redundancy that would suggest padding.
The surface covers failure patterns, flakiness, correlation, trends, screenshots, and consecutive failures well. Minor gaps exist—no tool to fetch individual run detail or list/search tests directly—but core triage and reliability-analysis workflows are covered.
Direct access to Cypress tests results and accessibility reports in your AI workflow.
Approved test intent, reviewed Playwright automation and run evidence, inside your editor.
Flaky test detection, root cause analysis, and fix suggestions for development teams.
Run, debug, and triage tests from your IDE using natural language, no dashboard switching, no manual data transfers. The TestMu AI (formerly LambdaTest) MCP Server is a single remote server exposing four tool suites: HyperExecute — analyze your project, generate YAML configs and test runner commands, then monitor jobs and sessions. Automation — pull a TestID's details plus command, network, and console logs into one chat for instant root-cause analysis. Includes mobile app upload. SmartUI — explain pixel, layout, DOM, and perceptual changes in a visual regression run, with context-aware React/HTML/CSS fixes. Accessibility — audit any public URL or a local React app against WCAG and get ready-to-apply remediation steps. Connects over https://mcp.lambdatest.com/mcp using OAuth 2.1 — no API keys in your config. One-click install in Cursor; works with Claude, GitHub Copilot, Cline, and any MCP client. Tests execute on the TestMu AI cloud: 3,000+ browsers and 10,000+ real devices.
