{"id":2278,"date":"2026-09-26T13:18:43","date_gmt":"2026-09-26T13:18:43","guid":{"rendered":"https:\/\/advancedcskills.com\/?p=2278"},"modified":"2026-09-30T14:15:45","modified_gmt":"2026-09-30T14:15:45","slug":"codex-101-for-construction-pms-and-vet-leaders-working-inside-your-project-folder","status":"publish","type":"post","link":"https:\/\/advancedcskills.com\/?p=2278","title":{"rendered":"Codex 101 for Construction PMs and VET Leaders: Working Inside Your Project Folder"},"content":{"rendered":"\t\t
\n\t\t\t\t
\n \t\t\tArticle<\/span>\n\t\t\t\t\t

Codex 101 for Construction PMs and VET Leaders: Working Inside Your Project Folder<\/h1>\n\t\t\t\t\t

Published 26 September 2026<\/p>\n\t\t\n<\/div>\n

\n \t\t\t

Codex is not just a chat box. It works inside a project folder on your computer, where the real work already lives: RFIs, variations, site photos, modules, mapping tables, learner evidence, templates and review notes. This article explains the idea, a five-level ladder for building up to it safely, and what has changed in OpenAI\u2019s tools since our guide was written.<\/p>\n\t\t\t\t\t

You do not need to code to use it. You do need to decide what it may read, what it may change, and who reviews the result.<\/p>\n\t\t\t\t\t

The key idea: the folder is the workspace<\/h2>\n\t\t\t\t\t

In a browser chat you copy pieces of your work into a window. With Codex you bring the AI to the work folder instead. It can read the files there, compare them, create new ones and follow standing instructions across sessions, using the permissions you give it.<\/p>\n\t\t\t\t\t

That is the difference from browser-based tools such as ChatGPT or Gemini in chat mode, which mainly talk with you. It matters whether you manage variation registers, SWMS documents and apprentice logbooks, or a corpus of PDFs and interview transcripts. The underlying capability is the same: read what is in the folder, work across several files, and produce something you can review.<\/p>\n\t\t\t\t\t

Two familiar scenarios<\/h2>\n\t\t
\n
\n \t\t\tQuantity surveyor<\/span>\n\t\t\t\t\t

Load RFIs, variations, site instructions, subcontractor quotes, daywork sheets, photos, claims, email threads and the contract schedule. Ask Codex to trace what changed, what it affected, what evidence supports it, and where the cost impact may sit.<\/p>\n\t\t\t\t\t

What it is doing:<\/strong> building an evidence trail of event, source documents, dates, cost category, missing records, assumptions and items needing QS review.<\/p>\n\t\t\t\t\t

The human still owns:<\/strong> entitlement, valuation, the contract position, and anything submitted commercially.<\/p>\n\t\t\n<\/div>\n

\n \t\t\tVET compliance<\/span>\n\t\t\t\t\t

Load units of competency, assessment tools, mapping documents, training and assessment strategy (TAS) files, learner samples, validation records, trainer credentials and existing modules. Ask Codex to map coverage, identify gaps, flag duplicate coverage and prepare a review pack.<\/p>\n\t\t\t\t\t

What it is doing:<\/strong> building a coverage map of requirement, module location, assessment task, evidence source, weak or duplicate coverage, and review questions.<\/p>\n\t\t\t\t\t

The human still owns:<\/strong> compliance judgement, assessment decisions, validation, and RTO accountability.<\/p>\n\t\t\n<\/div>\n\n<\/div>\n\t\t\t

The professional boundary:<\/strong> Codex can prepare the evidence trail. It does not own entitlement, valuation, safety sign-off, assessment judgement, compliance approval or trainer competency decisions.<\/p>\n\t\t\t\t\t

The five levels<\/h2>\n\t\t
\n \t\t\tLevel 1<\/span>\n\t\t\t\t\t

The Prompt User<\/h3>\n\t\t\t\t\t

What it is:<\/strong> you open Codex and ask isolated questions. It suits quick drafting, summarising and plain-English explanation, such as a clearer email about a delay or a simpler learner instruction.<\/p>\n\t\t\t\t\t

Why people stall:<\/strong> every prompt starts from zero. Codex has no access to the folder, project history, template or evidence.<\/p>\n\t\t\t\t\t

Move up:<\/strong> open the actual project folder and give Codex a real task with context.<\/p>\n\t\t\n<\/div>\n

\n \t\t\tLevel 2<\/span>\n\t\t\t\t\t

The Folder Worker<\/h3>\n\t\t\t\t\t

What it is:<\/strong> Codex can now see the job pack, not just your question. Give it the relevant folder and a clear definition of done: what to read, what to ignore, what output you want, and what must be flagged for review. A QS might ask for a draft change-event register with document references and missing evidence. A VET leader might ask for a draft coverage map and compliance questions.<\/p>\n\t\t\t\t\t

Why people stall:<\/strong> they dump files into a folder and hope. Codex still needs instructions, boundaries and output rules.<\/p>\n\t\t\t\t\t

Move up:<\/strong> ask Codex to trace every claim back to source documents. Evidence mapping is where the value starts to build.<\/p>\n\t\t\n<\/div>\n

\n \t\t\tLevel 3<\/span>\n\t\t\t\t\t

The Evidence Mapper<\/h3>\n\t\t\t\t\t

What it is:<\/strong> Codex compares, traces and groups, and shows its working. It links claims to source documents, shows gaps, and separates evidence from assumptions. Good output includes source references, confidence levels, missing evidence, contradictions, assumptions and items that need a human decision. The quality test is simple: a table is only useful if it leads back to the evidence.<\/p>\n\t\t\t\t\t

Why people stall:<\/strong> they accept the table without checking the trace.<\/p>\n\t\t\t\t\t

Move up:<\/strong> write standing instructions, and turn the mapping task into a repeatable workflow.<\/p>\n\t\t\n<\/div>\n

\n \t\t\tLevel 4<\/span>\n\t\t\t\t\t

The Workflow Builder<\/h3>\n\t\t\t\t\t

What it is:<\/strong> you stop explaining the same review method every time. A standing project brief (the AGENTS.md file) holds your rules, and reusable Skills hold common jobs: for a QS, an RFI-to-variation impact review or a progress claim support checker; for a VET leader, a unit-to-module mapper or a validation pack reviewer. A good Skill defines the input folder, evidence rules, output table, review flags, file naming and the human sign-off point.<\/p>\n\t\t\t\t\t

Why people stall:<\/strong> they get one good result but never save the process, so the next review starts from memory again.<\/p>\n\t\t\t\t\t

Move up:<\/strong> choose one low-risk recurring review and make it routine, with a clear human approval point.<\/p>\n\t\t\n<\/div>\n

\n \t\t\tLevel 5<\/span>\n\t\t\t\t\t

The Governed Operator<\/h3>\n\t\t\t\t\t

What it is:<\/strong> Codex becomes part of a controlled review system. The question is no longer \u201cCan Codex draft this?\u201d but what it is allowed to read, what it can draft, what it must never change, who reviews the output, and what audit trail you keep. Approvals, sandboxing, controlled connectors, Skills, automations, role-based access and review logs keep the workflow reviewable.<\/p>\n\t\t\t\t\t

Why people stall:<\/strong> trust. The tool is not the hard part. The hard part is deciding where human accountability sits.<\/p>\n\t\t\t\t\t

Best first routine:<\/strong> start with one valuable but low-risk review. Keep the first version boring, reviewable and human-approved.<\/p>\n\t\t\n<\/div>\n\t\t\t

What has changed since the guide<\/h2>\n\t\t\t\t\t

Our guide was last reviewed in May 2026, and OpenAI\u2019s tools have moved on since. Checked against OpenAI\u2019s documentation in September 2026:<\/p>\n\t\t\t\t\t