Hi ​@manivalidus,

Short answer is that Automation Anywhere hasn’t removed HTML support, but it has changed how technology types are chosen. In the current A360 Recorder:

• You can no longer explicitly select “HTML” at capture time—the Recorder auto‑detects the technology and shows it in Object properties instead. This is stated verbatim in the docs:
  “You do not have an option to select Java or HTML technology explicitly… the Recorder detects the technology automatically … the technology type is displayed within the Object properties.” 

• If you do force a specific technology, the only selectable options are MSAA, UI Automation, and UI Automation (COM)—HTML is not offered as a manual choice. 

Automation Anywhere consolidated recording into the Universal Recorder (and AISense). This integrated Recorder model replaced the older, technology‑specific approaches and serves as the common entry point. 

The Capture action replaced legacy Object Cloning from v11, further signaling the shift away from the older per‑tech workflows. 

Important nuance: HTML is still supported

• The Universal Recorder supports HTML applications (and still exposes HTML‑specific properties like DOMXPath, HTML tag, HTML name/ID, etc., under Object properties after capture). The docs list “HTML applications” under supported application types for the Universal Recorder. 
• The “Using the Capture action” page documents DOMXPath and other HTML properties and even recommends using DOMXPath over Path when automating web apps—again, reinforcing that HTML is present but auto‑detected, not manually chosen

What You Can Do

1. Enable Native Fallback for DOMXPath

   • Go to Bot Editor → Advanced Settings → Package Settings → Recorder and enable Use native fallback.
   • This ensures multiple DOMXPaths are stored and validated for dynamic elements.

2. Use Universal Recorder Instead of HTML Tech Type

   • Universal Recorder still captures HTML-based apps but under a unified approach.
   • Check if the DOMXPath property appears in the captured object properties.

3. Manual DOMXPath Handling

   • If Recorder doesn’t expose DOMXPath, you can:
     • Use the Run JavaScript action to interact with elements directly.
     • Or dynamically build XPath using variables in loops (e.g., //div[@id='container']/div[$Counter$]).

I would recommend registering for the below upcoming session.

Improved Self-Healing Automation with Vision Model-Powered Generative Recorder, Wed, 8 Oct. 2025 at 08:30, America/Los_Angeles | Community

Hello ​@Padmakumar 

Thanks for the response.

I don’t see this from my portal:

What You Can Do
Enable Native Fallback for DOMXPath

Go to Bot Editor → Advanced Settings → Package Settings → Recorder and enable Use native fallback.
This ensures multiple DOMXPaths are stored and validated for dynamic elements.
Use Universal Recorder Instead of HTML Tech Type

Universal Recorder still captures HTML-based apps but under a unified approach.
Check if the DOMXPath property appears in the captured object properties.
Manual DOMXPath Handling

If Recorder doesn’t expose DOMXPath, you can:
Use the Run JavaScript action to interact with elements directly.
Or dynamically build XPath using variables in loops (e.g., //div[@id='container']/div[$Counter$]).

Hello ​@manivalidus, I was thinking the new MV3 (Manifest version 3) extension could be the problem, so I rolled back my Recorder package to the same version you’re using and tried capturing. 

It captured correctly with HTML technology.

This leads me to believe there is something wrong with the extension in your browser.

What I would recommend is to:

1. Completely close your browser AND all tabs and windows, then relaunch just one browser window and try again. I know there are people who never shut down their browser or shut down their computer.  😊
2. If that doesn’t work, try removing the extension, uninstalling the Bot Agent, and reinstalling the Bot Agent. That should reinstall the browser extension. We need the MV3 browser extension with current versions of Chrome.
3. Try a fresh automation with the latest package version. In my control room, I see that my latest package version is 5.0.5-20250903-100026. You may not have this late a version, but try the latest one installed on your control room.
4. Try a standard web page, one without complexities. See if it captures properly.

Give those things a try and let us know!

Hello ​@Padmakumar 

Thanks for the response.

I don’t see this from my portal:

 

What You Can Do
Enable Native Fallback for DOMXPath

Go to Bot Editor → Advanced Settings → Package Settings → Recorder and enable Use native fallback.
This ensures multiple DOMXPaths are stored and validated for dynamic elements.
Use Universal Recorder Instead of HTML Tech Type

Universal Recorder still captures HTML-based apps but under a unified approach.
Check if the DOMXPath property appears in the captured object properties.
Manual DOMXPath Handling

If Recorder doesn’t expose DOMXPath, you can:
Use the Run JavaScript action to interact with elements directly.
Or dynamically build XPath using variables in loops (e.g., //div[@id='container']/div[$Counter$]).

 

I think this requires a license to be allocated to the user. Now, these licenses are the suite of an on-demand license type. You will need to talk to your AA partner manager for this allocation. Sharing the link here Intelligent Automation - Automator AI Base.

I think this requires a license to be allocated to the user. Now, these licenses are the suite of an on-demand license type. You will need to talk to your AA partner manager for this allocation. Sharing the link here Intelligent Automation - Automator AI Base.

​@Padmakumar, I think we’re getting farther away from the original question. If I understand correctly, ​@manivalidus was previously able to capture objects and have it use the HTML technology, but is unable to do so now. This isn’t about running the automation and it failing to find an object.

​@manivalidus, were you able to try my troubleshooting steps above?

Even I faced the same behavior even after having the MV3 AA extension with the same version of Chrome. Even though rolling back the Recorder package could likely fix the issue, I am seeing this as a bit of a weird approach, as the latest updates should provide additional functionality, and at the same time, they shouldn't deviate from what it was capable of doing in its earlier stage; hence, my comment.

​@Padmakumar You’re correct. Our software is constantly evolving and improving. One thing we don’t have control over, however, is the user’s computer. New software, new policies, new firewall restrictions, new browser restrictions (e.g., MV3)… all these can affect the user’s experience. We can both agree that this is not a wide-spread problem, so it’s very likely something with the user’s computer, not anything they need to upgrade with our software.

Since ​@manivalidus hasn’t responded with the troubleshooting steps, I think we can leave this post alone. They may have found the solution themselves which could be as simple as rebooting their computer and just not replied to this post.

​@manivalidus if you found any of the suggestions helpful, please mark a best answer. Or tell us how you resolved the issue - your post could help out a lot more people in the community.