Resource search

Search DataDrivenAEC

Open the full search page →

Trunk Tools

AI agents for construction documentation

AI-driven platform that organizes unstructured project data (specs, RFIs, drawings, schedules) and automates admin workflows with AI agents including TrunkText, TrunkSubmittal, and TrunkReview.

$70M raisedGrowth Stage
Founded 2021 · Suffolk Construction, Gilbane, Charps, AMLI Residential (exact total customer count not disclosed)
Website →← All Tools

Why This Tool Exists

The Problem

Construction teams waste 5-10 hours per week manually digging through project documents, RFIs, specifications, and drawings to find answers to routine questions. Project managers struggle to identify inconsistencies between documents and missing information during submittal reviews, leading to costly rework and schedule delays. Drawing revisions are manually reviewed, making it easy to miss changes.

The Solution

AI agents automatically organize project data and answer questions instantly with cited sources. TrunkSubmittal identifies missing, conflicting, and noncompliant information across documents. TrunkReview uses vision language models to detect marked and unmarked drawing changes automatically. Teams can ask natural language questions and get answers in seconds instead of hours spent hunting through folders.

How You Use It

Delivery Method
SaaSMarketplace Integration
Integrations
Procore (recent API access terminated by Procore in Sept 2025)Autodesk BuildAutodesk Construction CloudP6CMiCSharePointBoxDropboxEgnyteInEight
Project Phases
Construction Documentation, Construction Administration
Project Types
Commercial, Residential, Healthcare, Education, Infrastructure, Mixed-Use, Industrial, Hospitality, Institutional

Data Transparency

Exactly what this tool uses and how

Input
What it needs
Required:Project documents, Construction drawings, Schedules
Optional:RFIs, Submittals, Contracts, Meeting minutes, Procurement logs, Specifications, Safety logs
Formats:PDF, Native CAD formats, Text documents, Structured data from connected systems
Output
What you get
Format:Structured, searchable project data with AI-generated insights
Fields:Document Q&A responses with citations, Automated submittal discrepancy flags, Drawing revision analysis with visual overlays, Extracted project insights and recommendations, Schedule-linked documentation connections
Algorithm
How it works
Model:Proprietary vision language models and generative AI trained on construction documents
Accuracy:Not publicly specified
Privacy
How your data is protected
Retention:Project-based retention; customer data is customer-owned
Training:Uses customer data with explicit consent for model training (noted in privacy policy)
Compliance:SOC 2 Type II, CCPA compliance, Claims enterprise-grade security
API
Integration
Endpoint:Contact vendor for API access
Method:Not publicly documented

Use Cases

  • ·Instant answers to routine project document questions with source citations
  • ·Automated submittal compliance checking before submission to AHJ
  • ·Real-time drawing revision tracking and change notification
  • ·Schedule agent linking specifications and requirements to timeline
  • ·Field team mobile access to project documents and Q&A
  • ·Automatic RFI generation from discrepancies found in documents
  • ·Centralized searchable project knowledge repository

Pricing

Free
Demo available
Pro
Contact for pricing
Enterprise
Custom pricing for large contractors (contact sales)
Sources & Research NotesResearch date: 2026-02-01
Fields Checked (18)
problem, solution, deliveryMethod, yearFounded, maturityStage, fundingTotal, customerBase, disciplines, projectPhases, projectTypes, input-formats, output-structure, algorithm-approach, useCases, pricing-model, integrations-documented, privacy-location, compliance-soc2
Not Found (8)
api-documentation, api-rate-limits, algorithm-accuracy-metrics, data-retention-days, customer-count-exact, pricing-exact-amounts, procore-integration-status-active, data-training-explicit-policy-document
About this research: This record summarizes the cited sources and fields checked. A research date does not establish hands-on testing. Consult the original sources for current details and limitations.