Scripting

Rostyman lets you run JavaScript before and after every request. Scripts use the rm.* API — Rostyman's own scripting namespace (if you've used other API tools, it's a similar concept).

Script Types

Type	When it runs	Common uses
Pre-request Script	Before the request is sent	Fetch a token, set dynamic variables, modify headers
Test Script	After the response is received	Assert status codes, validate response body, set variables from response

Script Editor

Both script types use the Monaco editor (the same editor as VS Code) with:

• JavaScript syntax highlighting
• Auto-completion for rm.* API
• Error highlighting

Execution Environment

Scripts run in a sandboxed Node.js vm context in the Electron main process with:

• A 5-second timeout — scripts that run longer are killed automatically
• The rm.* API, plus pm.* / pw.* / tc.* / bru.* compatibility aliases so scripts imported from Postman, Hoppscotch, Thunder Client, and Bruno run unchanged
• A restricted set of JavaScript globals: JSON, Math, Date, RegExp, Array, Object, String, Number, the parse* / encode* / decode* helpers, and atob / btoa
• No require(), no file system, no network calls, no DOM, and no setTimeout / setInterval
• Isolated context — scripts cannot affect each other or the main process directly

Example: Select a Token Before Request

Scripts cannot make network calls. A common pattern is to read a token that was set beforehand (manually, by a login request's test script, or in an environment variable) and apply it as a request-scoped variable for auth:

// Pre-request Script
const token = rm.environment.get('accessToken')
rm.variables.set('activeToken', token)
// Now reference {{activeToken}} in your Bearer Token auth field

Example: Test Response

// Test Script
rm.test('Status is 200', () => {
  rm.expect(rm.response.code).to.equal(200)
})

rm.test('Response has user ID', () => {
  const body = rm.response.json()
  rm.expect(body).to.have.property('id')
  rm.expect(body.id).to.be.a('number')
})

// Save the ID for use in other requests
rm.environment.set('userId', rm.response.json().id)

Script Execution Order

When a request is inside folders and a collection, scripts execute in a specific order. Pre-request scripts run from outermost to innermost, and post-response scripts run in reverse.

1. Collection pre-request script
2. Root folder pre-request script
3. Sub-folder pre-request script(s) — outermost → innermost
4. Request pre-request script
         ← Request is sent →
5. Request post-response script
6. Sub-folder post-response script(s) — innermost → outermost
7. Root folder post-response script
8. Collection post-response script

Variable Cascade

Variables set by rm.environment.set(), rm.collectionVariables.set(), or rm.globals.set() in earlier scripts are available in later scripts in the chain. This means a collection script can set a token that folder and request scripts can read.

Where to Add Scripts

Level	How to open
Collection	Click collection name in sidebar → Scripts sub-tab
Folder	Click folder name in sidebar → Scripts sub-tab
Request	Request builder → Scripts tab (Pre-request / Post-response)

Test Results

After a request with test scripts, the Tests tab in the response panel shows:

✓  Status is 200
✓  Response has user ID

2 passed, 0 failed

Failed tests show the assertion message and expected vs actual values.

Code Snippets

The script editor includes a Snippets panel (click the icon in the toolbar) with ready-to-insert code examples for common tasks:

• Set/get environment variable
• Set/get collection variable
• Set/get global variable
• Log to console
• Parse JSON response
• Write assertions

Click any snippet to insert it at the cursor position.

Beautify

Click the Beautify button (or press Shift+Alt+F) in the script editor toolbar to auto-format your JavaScript. This uses the Monaco formatter and respects standard JS style.

AI-Assisted Scripting

Open the AI Assistant from the right panel and describe what you want the script to do. The AI can:

• Write a pre-request script from a natural language description
• Suggest test assertions based on the response body you paste
• Explain what an existing script does
• Fix errors in a failing script

Generated scripts are inserted directly into the script editor. You can edit them further before saving.