skills.rest
sgttomas
sgttomasCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

drawing-extract-page

Extract bounded drawing-page header data using crop-first VLM reasoning.

Updated Aug 27, 2026
Search Tags:#drawing#extraction#vlm#top-header#pfd#per-page
One-click install
npx skills add https://github.com/sgttomas/chirality-piping --skill drawing-extract-page
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: drawing-extract-page
Source: https://github.com/sgttomas/chirality-piping/tree/main/skills/drawing-extract-page
Command: npx skills add https://github.com/sgttomas/chirality-piping --skill drawing-extract-page

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Bounded per-page extraction of structured data from engineering drawings using crop-first, multiblock reasoning to provide precise results for top header data and drawing information.

Core Features & Use Cases

  • Bounded per-page extraction of top-header items (equipment blocks) and title-block data from a drawing page.
  • Supports PFD detailed targets and basic targets with field-level detail; stubbed types fail fast with clear messaging.

Quick Start

Dispatch a per-page extraction for a single drawing page image using DRAWING_EXTRACT with DRAWING_TYPE=PFD and extraction_target=top_equipment_header_basic for basic extraction or top_equipment_header_detailed for detailed extraction.

Frequently Asked Questions about drawing-extract-page

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract top header data from engineering PFD drawings?▼

To extract top header data from PFD drawings, dispatch per-page extraction using DRAWING_EXTRACT with DRAWING_TYPE=PFD and extraction_target=top_equipment_header_basic. This applies crop-first VLM reasoning to capture bounded equipment block data.

What is crop-first VLM reasoning for per-page drawing extraction?▼

Crop-first VLM reasoning is an extraction mechanism that crops bounded drawing-page regions before applying vision-language models. This multiblock reasoning approach provides precise per-page outcomes for structured top header and title-block data.

Can I extract detailed equipment block information from a PFD page?▼

Yes, you can extract detailed equipment block information by setting extraction_target to top_equipment_header_detailed. This targets field-level detail from bounded per-page drawing regions using crop-first multiblock VLM reasoning.

What extraction targets are supported for PFD drawing data extraction?▼

Supported PFD extraction targets include top_equipment_header_basic for basic extraction and top_equipment_header_detailed for field-level detail. Stubbed or unsupported extraction targets fail fast with clear messaging to ensure valid inputs.

What output format does per-page drawing extraction produce?▼

Per-page drawing extraction produces a single markdown_stub artifact containing the extracted structured data. This output captures bounded top-header items and title-block information validated against DRAWING_TYPE and EXTRACTION_TARGET inputs.

Related Skills

Explore More Data & Analytics Skills→
flyingtimesflyingtimes

kpi-analyst

Extract KPI indicators from Excel workbooks and generate per-sheet Markdown reports.

Community
Intermediate
robikscuberobikscube

sheet-splitter

Split bound construction drawing set PDFs into individual per-page sheet files.

Community
Intermediate
9aia9aia

nuxt-seo

Configure Nuxt SEO modules to generate robots.txt, sitemap.xml, OG images, and JSON-LD.

Official
Advanced
pmcdowallpmcdowall

segment-anything-model

Segment images into object masks using point, box, or mask prompts.

Community
Advanced