---
title: "UNC and Microsoft's WebXSkill Pushes Web Agents to 86% on Real Sites"
subtitle: "A Microsoft and UNC team gives web agents reusable skills that are both executable and inspectable, boosting task success by up to 12.9 points."
url: "https://alphasignal.ai/news/unc-and-microsoft-s-webxskill-pushes-web-agents-to-86-on-real-sites"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-04-14T00:00:00.000Z"
updated: "2026-08-23T23:07:03.638Z"
type: "Paper"
topics:
  - "Agents"
  - "Development"
subtopics:
  - "Computer Use"
  - "Tool Use"
  - "Code Agents"
  - "Code Agents"
upvotes: 2911
source_url: "https://arxiv.org/abs/2604.13318"
comments: 0
truncated: true
---

# UNC and Microsoft's WebXSkill Pushes Web Agents to 86% on Real Sites

> A Microsoft and UNC team gives web agents reusable skills that are both executable and inspectable, boosting task success by up to 12.9 points.

Web agents built on large language models can click buttons and fill forms, but they still fall apart on long, multi-page workflows. A new framework called [WebXSkill](https://arxiv.org/abs/2604.13318), from researchers at UNC Chapel Hill and Microsoft, gives agents a library of reusable skills they can either execute directly or follow step-by-step, closing what the authors call the grounding gap between plain-text workflows and opaque code.

## Why agents keep forgetting what they just learned

When a web agent successfully completes a checkout flow or navigates a complex admin panel, the procedural knowledge embedded in that trajectory is often discarded. The next time a similar task appears, the agent re-plans the entire sequence from scratch, wasting steps and inviting new hallucinations.

Prior work has tried to fix this with skills, and each approach compromises something. Workflow-based systems such as AWM represent skills as natural language instructions that guide planning but cannot be directly executed, forcing the agent to translate each instruction into concrete browser actions and reintroducing grounding errors. Code-based approaches like SkillWeaver and WALT go the other direction, grounding skills in executable scripts but deploying them as opaque black-box calls without step-level guidance, leaving the agent unable to understand, adapt, or recover when execution fails mid-way.

![Comparison of textual workflow skills versus WebXSkill executable skills](https://arxiv.org/html/2604.13318v1/teaser.png) 

## Executable skills with a voice

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [WebXSkill: Skill Learning for Autonomous Web Agents](https://arxiv.org/abs/2604.13318)
- [aiming-lab/WebXSkill GitHub repository](https://github.com/aiming-lab/WebXSkill)
- [browser-use: browser automation for AI agents](https://github.com/browser-use/browser-use)