Knowledge Corpus Inventory
Every document, template, SOP, example, and decision rule in the chosen knowledge area, catalogued at a high level.
Proof
Shows what AI would actually find if it searched the area.
AI tools that search, summarise, draft from, or reuse internal knowledge need a clean corpus. One month to deeply audit one bounded knowledge area and identify what must change before AI touches it.
Before an AI assistant searches a knowledge area, you need to know whether that specific area is safe to search. This is not a broad business-wide knowledge review. It goes deep on one selected area.
This service audits one bounded knowledge area: one practice area, one service line, one project type, one recurring delivery context, or one document corpus. It assesses whether that corpus is usable by staff and safe for AI retrieval or reuse.
I identify where the documents, templates, SOPs, examples, and decision rules in that corpus actually live. I sort what is current and trusted from what is stale, duplicated, restricted, missing, or unsafe. The output is a clear cleanup plan and structure so that when AI tools search or reuse that knowledge area, they find the right material.
Every document, template, SOP, example, and decision rule in the selected corpus is reviewed, classified, and assigned a status. Nothing is assumed to be retrieval-ready until I have seen it.
This layout is for explaining a service as a repeatable operating shape: who it is for, how the work moves, and what the client leaves with.
You are planning to use AI tools that will surface internal documents, SOPs, templates, examples, or guidance. You need to know the state of that selected corpus before it becomes AI search results.
One practice area, service line, project type, or recurring delivery context depends on templates, examples, SOPs, project files, and staff judgement that need to be made clear before AI can reuse them.
Client documents, sensitive project files, and restricted material sit in the same systems as general content. AI access needs boundaries before it gets broad access.
Select one bounded knowledge area with leadership. Define the corpus, retrieval use case, systems in scope, and classification criteria I will apply.
Week 01Map every document, template, SOP, example, and decision rule in the selected corpus. Identify where items live, who owns them, and when they were last updated.
Week 02Assign every item a status: trusted and current, stale and needs update, duplicated, restricted, or missing entirely.
Week 03Present the full audit findings and a specific clean-up brief: what to fix, what to remove, what to write, and in what order.
Week 04Every document, template, SOP, example, and decision rule in the chosen knowledge area, catalogued at a high level.
Shows what AI would actually find if it searched the area.
The documents and sources that are current, authoritative, and safe to use.
Separates usable knowledge from accidental clutter.
Material that is outdated, duplicated, or conflicting, and what to do with it.
Prevents old guidance being treated as current.
Material that must be excluded from AI access and why.
Keeps confidential material out of the wrong context.
A practical folder or taxonomy structure that makes the knowledge easier to find and maintain.
Turns scattered knowledge into something staff and AI can navigate.
What to fix first, in order.
Turns the audit findings into a workable action list.
A plain-English assessment of whether the knowledge area is ready for AI to search, summarise, draft from, or reuse, and what needs to change before it is.
Connects the audit directly to AI readiness.