reworkd/tarsier
Vision utility library that tags interactable webpage elements with IDs for LLMs to issue commands like CLICK [23].
Not currently ranked — collecting fresh signals.
star history
Tarsier provides visual perception capabilities for web interaction agents by tagging buttons, links, and input fields with brackets and IDs, creating a mapping between DOM elements and identifiers that LLMs can reference. It integrates with Playwright and Selenium to capture webpage state, supports OCR for visual structure understanding, and works alongside GPT-4V for multimodal page understanding. The library is published as a PyPI package and is used by Reworkd to power autonomous web agents across thousands of real tasks.
Frequently asked
- What is reworkd/tarsier?
- Vision utility library that tags interactable webpage elements with IDs for LLMs to issue commands like CLICK [23].
- Is tarsier open source?
- Yes — reworkd/tarsier is open source, released under the MIT license.
- What language is tarsier written in?
- reworkd/tarsier is primarily written in Jupyter Notebook.
- How popular is tarsier?
- reworkd/tarsier has 1.8k stars on GitHub.
- Where can I find tarsier?
- reworkd/tarsier is on GitHub at https://github.com/reworkd/tarsier.