Skip to content

Comments

TestDino MCP#3362

Open
TestDinoHQ wants to merge 2 commits intomodelcontextprotocol:mainfrom
TestDinoHQ:main
Open

TestDino MCP#3362
TestDinoHQ wants to merge 2 commits intomodelcontextprotocol:mainfrom
TestDinoHQ:main

Conversation

@TestDinoHQ
Copy link

Add TestDino MCP to the MCP Server Registry as a production-grade testing and debugging server for Playwright and web test suites. TestDino MCP connects AI agents to the TestDino platform so they can explore test runs, analyze failures, work with manual test cases, and perform AI-assisted debugging using natural language.

Publishing Your Server
Published via the MCP Server Registry following the quickstart guide.

Server Details
Server: testdino-mcp (TestDino MCP – AI-native test intelligence server)

Changes to: tools (12 tools for test runs, test cases, debugging, and manual test management), documentation links, and metadata for registry discovery.

Motivation and Context
Modern QA teams are moving from static dashboards to AI-assisted workflows, but most MCP servers focus on infra/DevOps and not on test intelligence. TestDino MCP fills this gap by exposing rich, structured test data (runs, cases, artifacts, history) via MCP, allowing LLM clients to debug flaky tests, surface patterns, and manage manual test assets directly from editors like Cursor and Claude Code. This makes automated testing data actionable from within AI coding environments and shortens the feedback loop between failures and fixes.

How Has This Been Tested?
Validated with MCP-compatible clients (Cursor, Claude Desktop/Code) using npx testdino-mcp and global install flows.

Tested scenarios:

  • Health checks and PAT validation via health.

  • Listing and filtering test runs by branch, author, time window, environment, and commit using list_testruns.

  • Fetching detailed run data and individual test case details (errors, traces, logs, artifacts) via get_run_details and get_testcase_details.

  • Debugging flaky and failing tests with debug_testcase (historical aggregation, failure patterns, browser-specific issues, and AI-ready prompts).

  • Managing manual test cases and suites (list, create, update) via list_manual_test_cases, create_manual_test_case, update_manual_test_case, and list_manual_test_suites.

Breaking Changes

No breaking changes. This is a new server entry in the registry and does not require existing MCP clients to change behavior. Existing users can opt in by adding the TestDino MCP config (command + TESTDINO_PAT) to their MCP client settings.

Types of changes

  • Bug fix (non-breaking change which fixes an issue)

  • New feature (non-breaking change which adds functionality)

  • Breaking change (fix or feature that would cause existing functionality to change)

  • Documentation update

Checklist

  • I have read the
    MCP Protocol Documentation

  • [ ]

  • [ ]
    My changes follows MCP security best practices (uses PAT via TESTDINO_PAT, no secrets logged, PAT scoped to user’s TestDino permissions).

  • [ ]
    I have updated the server's README accordingly (installation, tools reference, and client examples).

  • I have tested this with an LLM client (Cursor / Claude Desktop / Claude Code).

  • My code follows the repository's style guidelines (Node 18+, standard MCP server patterns).

  • [ ]
    New and existing tests pass locally (basic health and tool invocation smoke tests).

  • [ ]
    I have added appropriate error handling (PAT validation, TestDino API errors, input validation for tool parameters).

Additional context

  • Server is distributed via npx testdino-mcp, global npm install -g testdino-mcp, or project install.

  • One PAT gives access only to organizations and projects the user already has permissions for in TestDino, so authorization is delegated to the platform.

  • The debug_testcase tool is optimized for AI workflows: it returns a compact, structured summary plus a ready-to-use debugging prompt that LLM clients can feed into their own reasoning loop.

Added testdino mcp
Added TestDino MCP here
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant