Content Scripts

Content Scripts

Content scripts run in the context of web pages in an isolated world. This allows multiple content scripts from various extensions to coexist without conflicting with each other's execution and to stay isolated from the page's JavaScript.

Use cases:

Adding a single content script


Since Plasmo's default Typescript configuration treats all source files as modules, if you don't have any imports or exports in your code, you'll have to add an export {} line at the start of your file. (You will see this warning when creating your first content script!)

Create a content.ts file that exports an empty object and hack away!

export {}
  "You may find that having is not so pleasing a thing as wanting. This is not logical, but it is often true."

Reload your extension, open a web page, then open its inspector:


See with-content-script (opens in a new tab) for a full example.

Adding multiple content scripts

Create a contents directory for multiple content scripts, and add your content scripts there. Make sure their names describe what they do!

See with-many-content-scripts (opens in a new tab) for an example.


Sometimes, you'll want to run a content script on certain pages. You can provide a custom content script configuration by exporting a config object from your content script:

import type { PlasmoCSConfig } from "plasmo"
export const config: PlasmoCSConfig = {
  matches: ["<all_urls>"],
  all_frames: true

Working with this configuration object is a breeze thanks to the exported PlasmoCSConfig type 🥳.

To learn more about the config and each property, check out Chrome's official documentation (opens in a new tab).

Injecting into the main world

You must inject code into the main world if you'd like to modify the window object from your content script. It's not currently possible to declaratively inject content scripts.

Instead, Chrome offers a chrome.scripting.executeScript API that lets you inject content scripts into the main world. First, add the scripting permission in your package.json's 'manifest.permissions' array:

  "manifest" : {
    "permissions": ["scripting"]

Then, inject your content script into the main world by calling chrome.scripting.executeScript from your background service worker:

    target: {
      tabId // the tab you want to inject into
    world: "MAIN", // MAIN to access the window object
    func: windowChanger // function to inject
  () => {
    console.log("Background script got callback after injection")

For the func key, you can pass in a TS function from your project. It will be transpiled into JS when your extension bundles.


Any import inside the injected func will not be resolved and bundled at the moment. This is because the entire function must be self-contained and scoped entirely within the executeScript call, and we have yet to figure out a good bundling strategy to support this.

See with-scripting (opens in a new tab) for an example.

Fetching external API and CORS

Because content scripts run within the context of a web page, they are subject to the same-origin policy. To mitigate CORS restrictions, you can use the Plasmo Messaging API to proxy the request via your background service worker.

Importing resources

To import external assets into your content script, you can use the url: scheme:

import myFile from "url:./path/to/my/file/something.js"

The url: scheme will automatically resolve the something.js asset and add it to the web_accessible_resources declaration in the built bundle. The above myFile variable will be a string containing the URL to the asset:

> console.log(myFile)
chrome-extension://<your chrome ext id>/something.eb20bc99.js?1656000646313

Alternatively, you can use the data-base64 or the data-text scheme to import and embed the asset directly into your code. For small assets, these schemes should work well.

Last updated on February 5, 2023