specification-extractor_skill

This skill extracts structured CSI specification data from construction documents, identifying sections, products, submittals, and standards for procurement.
  • Python

2.6k

GitHub Stars

4

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill specification-extractor

  • _meta.json492 B
  • claw.json550 B
  • instructions.md1.2 KB
  • SKILL.md14.8 KB

Overview

This skill extracts structured data from construction specification documents and converts narrative spec text into actionable records. It parses CSI MasterFormat section headers, separates Part 1/2/3 content, and identifies manufacturers, product requirements, standards, and submittal obligations. The output supports estimating, procurement, compliance checks, and submittal tracking.

How this skill works

The extractor reads PDF text, detects CSI section headers and PART divisions, and parses article-level items into structured parts. It identifies manufacturers and product lines in Part 2, locates submittal requirements in Part 1 using keyword mappings, and scans for referenced standards with regex patterns. Results are returned as a SpecExtractionResult containing sections, products, submittals, and referenced standards, plus optional generated reports and logs.

When to use it

  • Preparing estimates from project specifications to quickly identify scope and product requirements
  • Creating procurement schedules or product lists from specification documents
  • Generating a submittal log to track required shop drawings, samples, and product data
  • Validating spec compliance by extracting and reviewing referenced standards
  • Archiving and indexing large numbers of specification PDFs for search and analytics

Best practices

  • Provide clean, machine-readable PDFs (text-based rather than scanned images) for best extraction results
  • Confirm CSI-style section headings are present or pre-process headings to match MasterFormat patterns
  • Review and normalize manufacturer names and product fields after extraction to match your procurement catalog
  • Tune or extend standard and submittal keyword patterns for project-specific terminology
  • Validate extracted submittal timing and copy counts against contract requirements before issuing logs

Example use cases

  • Run a batch of project specs to build a vendor/product shortlist for procurement and estimating
  • Auto-generate a submittal log for the general contractor showing required shop drawings and samples
  • Produce a standards register that lists all referenced ASTM, ANSI, UL, and NFPA items across specs
  • Populate an internal BOM or product schedule from Part 2 product descriptions for materials planning

FAQ

The implementation expects text-extractable PDFs. Scanned images require OCR before feeding to the extractor.

How accurate is manufacturer and product extraction?

Extraction is rule-based and reliable for well-structured specs but often requires manual review and normalization for inconsistent naming and complex product descriptions.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational