agentsclimarketplace

Do wdr visual resolver

Skill d-oit/do-web-doc-resolver/.agents/skills/do-wdr-visual-resolver

LLM-ready web documentation resolver: Python cascade skill + web + Rust CLI (wdr) with semantic cache, multi-provider routing, and quality synthesis

Install
npx -y skills add d-oit/do-web-doc-resolver --skill do-wdr-visual-resolver

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Visual URL resolution skill using CLIP screenshot embeddings to handle scanned PDFs, image-heavy layouts, and JS-heavy SPAs.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

1.3 KB, as published. Nobody here has run it

Visual Resolver Skill

Extends the Web Doc Resolver cascade with a visual provider based on CLIP screenshot embeddings.

When to use this skill

Activate this skill when text-based extractors (Jina, Firecrawl, etc.) fail to resolve a URL, particularly for:

  • Scanned PDFs (raster text)
  • Single Page Applications (SPAs) returning skeleton HTML
  • Infographic-heavy or chart/table-only documents
  • Complex multi-column academic layouts

Prerequisites

Install visual resolution dependencies:

pip install torch torchvision torchaudio
pip install git+https://github.com/openai/CLIP.git
pip install playwright
playwright install chromium

Integration

This skill is designed to be integrated into the scripts/_url_resolve.py cascade.

Usage

from .visual_resolver import VisualResolver

resolver = VisualResolver()
if resolver.is_available():
    result = resolver.resolve("https://example.com/scanned-pdf", "technical architecture diagram")

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.