PEAT Document Assessment System

challenge

Challenge 1

brief

PEAT Document Assessment System developed an interactive assurance evidence assessment solution that applies large language models to analyse project documentation, score maturity, and surface assurance evidence and gaps aligned to recognised governance frameworks.

Please be aware that this content was generated follwing an automated review so may not be perfectly accurate; refer to the original challenge brief and team files for authoritative information

key outcomes

Expected to reduce time spent on manual evidence searching, improve consistency and transparency of assurance reviews, and enable earlier identification of evidence gaps to support faster, higher-quality assurance decisions.

important files

• enhanced_peat_interface.html: Interactive web interface for document selection, assurance scoring, and visualisation of evidence coverage across categories.
• export_lmm.json: Structured output of LLM-identified assurance evidence, categories, and confidence scores extracted from project documents.
• knowledge_graph.json: Knowledge graph linking documents, evidence items, categories, and entities to support explainable assurance insights.
• Prompt Engineering for Hackathon.docx: Documented prompt patterns used to guide LLM-based assurance assessment against defined standards.

details

team: PEAT Document Assessment System members: Sam Sheldon, Warwick Gross, Erin Hewitt topics: solution-centre, hack25, challenge1, large-language-models, natural-language-processing, html, javascript, d3-js, project-assurance, automation, evidence-management, llm, assurance-maturity technologies: Large Language Models, Natural Language Processing, HTML, JavaScript, D3.js