Use Cases

Once you’ve configured an AI agent with Bruno, you can use it to automate common API development tasks. The following use cases apply to any supported agent (Cursor, VS Code with Copilot, Codex, Claude); specific examples use Cursor-style prompts where relevant.

Common Use Cases

• Generate Bruno collections – Create API collections from your backend source code with requests, tests, and documentation.
• Write scripts and tests – Use the agent to write pre-request scripts, test scripts, and assertions. Reference Bruno’s JavaScript API Reference and Bru Lang in your instructions so the agent produces valid Bruno files.
• Automate workflows – Get help with CI/CD pipelines, environment setup, and collection organization.

Example: Create a complete Bruno collection from your backend API

If you want to create a new Bruno collection for your backend API endpoints from source code, you can use a prompt like:

Create Bruno collection for my backend API with relevant test cases and write docs for each request at request level.

The agent will analyze your source code and generate a Bruno collection. Your project may look like this:

my-project

.cursor

rules

instructions.md

src

routes

app.js

server.js

bruno-collection

environments

local.bru

prod.bru

auth-api.bru

product.bru

user.bru

bruno.json

package.json

.gitignore

Example: Create a CI/CD pipeline for API tests

You can ask the agent to generate a CI/CD pipeline for your API tests using Bruno CLI. For example, with Cursor you can use:

Create fully functional CI/CD pipeline for my API tests using Bruno CLI to generate test reports.

The agent can generate a workflow file (e.g. GitHub Actions) in your project:

my-project

.cursor

rules

instructions.md

.github

workflows

api-tests.yml

src

bruno-collection

package.json

.gitignore

Example: Write automated test cases

You can ask your AI agent to write test cases for each API in your Bruno collection, including different edge cases.

Prompt

Use a prompt like:

Write test cases for each API in my Bruno collection. Include happy-path checks and different edge cases (error responses, missing fields, invalid data, response time, headers).

The agent generates test scripts. Bruno uses the Chai library for assertions—all Chai expect syntax works in Bruno tests. The agent can produce tests such as:

Status codes and response body (happy path):

test("should return success", function () {
  expect(res.getStatus()).to.equal(200);
});

test("should return user data", function () {
  const body = res.getBody();
  expect(body).to.have.property("id");
  expect(body.name).to.equal("John Doe");
  expect(body.email).to.contain("@example.com");
});

Edge cases (error responses, conditional logic):

test("should validate response based on status", function () {
  const status = res.getStatus();
  const body = res.getBody();

  if (status === 200) {
    expect(body).to.have.property("data");
    expect(body.data).to.not.be.empty;
  } else if (status === 404) {
    expect(body).to.have.property("error");
    expect(body.error.message).to.contain("not found");
  } else {
    throw new Error(`Unexpected status code: ${status}`);
  }
});

test("should validate error response structure", function () {
  const status = res.getStatus();
  if (status >= 400) {
    const body = res.getBody();
    expect(body).to.have.property("error");
    expect(body.error).to.have.property("message");
    expect(body.error.message).to.be.a("string");
  }
});

For more test patterns (headers, response time, arrays and loops, saving variables for the next request), see the Testing introduction.