JevMade field notes / Architecture and workflow guide

How to build an AI tool that finds and removes ads

This guide explores a test project that uses an AI tool to spot and remove website ads. It shows how standard software handles the actual removal while the AI only decides if something is an advertisement.

Original by realZachiClassification

Listen to this guide

JevMade’s plain-English explanation

0:00 /

AI narration

Credits and license

“typesafe-adblock” by realZachi. Read the original source.

This expanded guide is an AI-narrated adaptation of the source’s essential explanation, examples and caveats, not a word-for-word reading. The synthetic voice does not imitate the author or imply their endorsement.

License: MIT

MIT License

Copyright (c) 2026 Mahmoud Ali Khan

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Our summary

This project is a custom web browser tool that removes advertisements from web pages. A person might use this to see how an AI tool can judge parts of a website as you browse, rather than using it as a complete replacement for standard ad blockers.

Standard software first searches the page for likely ads based on labels like the word sponsored. It sends these options to Jev, an AI tool that chooses from options rather than writing an answer. If Jev is confident an item is an ad, the software removes it.

This setup requires a private access code called an API key, which costs money for every page scanned. It will miss some ads and cannot block hidden tracking software. It is mostly useful for people who want to test how AI handles simple choices safely.

Key takeaways

Use standard software to find likely ads before paying an AI assistant to check them.
Keep the actual removal rules in your software so the AI cannot delete random items.
Test the tool with a highlight mode to catch mistakes before letting it delete anything.

More guides →See the experiments →