Xplorer - an easy way to explore API's

APIs are all around us. When delivering software, there is a need for your team to make ad-hoc calls to APIs and eyeball the responses without writing code. Empowering your teams with an easy way to explore APIs - can elevate the overall quality of your end-user experience.

The challenge is that APIs are designed for computers to talk to each other. The user-interface, when exploring APIs, should abstract away the underlying complexity of HTTP and JSON as far as possible. This is where a well-designed API exploration tool can make a huge difference.

Karate is an industry leading open-source test automation platform combining API testing, API performance testing, API mocks & UI testing. Use by 500+ companies including 76 of the Fortune 500, are a testimony to Karate making test automation simple and fun. We did this by making it exceptionally simple to write tests.

We launched Xplorer in 2023 giving designers, developers, and product-owners - the ability to explore APIs, import/export different API data sources, environment switching and more.

  • Install and run locally (no cloud)
  • API chaining
  • Environment switching
  • JSON support
  • Before / After scripts
  • Label API calls    
  • Input Switching    
  • Mask Secrets
  • Readable Flows
  • Call APIs on localhost or private network
  • Export to Karate Test 
  • Import Postman Collections, cURL, HAR, OpenAPI, Swagger and make API calls (Premium feature)

Xplorer embeds an MCP server for ease of use and is easy to integrate with any LLM or IDE that can talk to MCP servers. This means that you are in control over your data or in other words, you "Bring Your Own LLM", so even a local LLM or the one approved for use for your organization will work, if it supports MCP.

The following core capabilities are exposed:

  • xplorer_script - let an LLM remote control Xplorer and do the hard work of creating chained flows with variables and dynamic requests
  • http_client - a scriptable HTTP client that allows the LLM to make calls and re-try input parameters until they work.
  • api_spec_utilities - able to read an OpenAPI or Swagger spec from the file-system or URL and work with it
  • http_mock_server - (experimental) spin up a mock server on the fly and program it to handle state and dynamically respond to incoming HTTP requests

When you combine all the four above, you can get some very useful results. For example:

  • upload a requirements document, make a logical sequence of HTTP calls and add them to Xplorer
  • refer to an OpenAPI specification and ask the LLM to create HTTP calls and add them to Xplorer
  • create a mock-server that replicates some functionality of an existing API
  • make sure that authentication, and setting variables and dynamic request parameters works in Xplorer
  • ask the LLM to explain the sequence currently open in Xplorer

Refer here for detailed features.

Get Started at here and give us your feedback at info@karatelabs.io . 


Happy Xploring!