System: MWMS
Document Type: Standard
Authority Level: MCR Source Of Truth
Status: Draft For MCR
Primary Location: MCR
Future Operational Destination: HeadOffice Brain, AI Business Systems Brain, AI Manager, AI Employee Router, Brain Room, Course Absorption System, Content Brain, Offer Brain, Sales Brain, Creative Brain, Future AIBS Client Systems
Parent Page: HeadOffice
Owner: Martyn
Developer Boundary: No Development Action Authorized By This Page
Source Of Truth: MCR
Purpose
The purpose of this document is to define the MWMS Context Library Governance And Folder Map Standard.
This standard establishes how MWMS organizes, stores, separates, updates, archives, and governs context libraries, skills, project folders, client libraries, and reusable AI source files.
MWMS depends on context.
But context becomes dangerous when it becomes messy.
A poorly governed context library can cause AI Employees to:
read the wrong files
use outdated instructions
mix client information
duplicate source-of-truth files
apply the wrong offer context
resurrect retired language
confuse draft material with approved material
treat project files as permanent context
use stale customer language
produce inconsistent output across Brains
This standard exists to prevent context chaos as MWMS scales.
MWMS must not rely on random folders, scattered documents, duplicated files, or ungoverned context dumps.
Each serious offer, client, Brain, skill system, or AI workflow must have a clear folder structure and authority model.
Scope
This standard applies to all MWMS context libraries, offer libraries, client libraries, skill folders, project folders, source-of-truth files, and reusable AI context packs.
This includes:
HeadOffice Brain
AI Business Systems Brain
AI Manager
AI Employee Router
Brain Room
Course Absorption System
Newsletter Intelligence
Opportunity System
Content Brain
Offer Brain
Creative Brain
Sales Brain
Conversion Brain
Customer Brain
Research Brain
Affiliate Brain
Ads Brain
future AIBS client systems
This standard applies to:
internal MWMS context libraries
offer-specific context libraries
client-specific context libraries
AI Employee context folders
skill folders
project folders
archive folders
retired language files
raw source material folders
output asset folders
audit folders
This standard does not authorize development work, plugin changes, Supabase changes, WordPress changes, automation wiring, or changes to M’s active build areas.
Core Definition
A Context Library is a governed collection of source files that AI Employees may use to understand a business, offer, client, Brain, workflow, or task.
A Folder Map is the approved structure that defines where those files live and how they are separated.
Context Library Governance defines:
where source files live
which folder is source of truth
which files are active
which files are archived
which files are project-only
which files belong to a client
which files belong to MWMS
which skills may read which context
which files must not be duplicated
which files must be updated when things change
which files must be retired or parked
Core Principle
The core principle of this standard is:
Context must be easy to find, hard to confuse, and safe to reuse.
A context file should have one clear source-of-truth location.
A client library should stay separate from MWMS internal context.
A project folder should not become permanent memory by accident.
A skill should read from the right library, not carry duplicated context inside itself.
MWMS must design folders so that future AI Employees, humans, and client systems can understand what each file is, where it belongs, and whether it is safe to use.
Primary Folder Layers
MWMS uses three primary folder layers for context-based work.
- Context Library
The Context Library stores durable source-of-truth context.
Examples:
Offer Context Library
Client Context Library
MWMS Brain Context Library
Business Context Library
AI Employee Context Pack
This folder contains files that should remain useful beyond one task or project.
- Skills Folder
The Skills Folder stores reusable procedural skills.
A skill defines how to do work.
It does not replace context files.
Examples:
Course Absorption Skill
Newsletter Signal Filtering Skill
Offer Evaluation Skill
Voice Checker Skill
Client Report Drafting Skill
Developer Handoff Precision Skill
Skills should read from approved context where possible.
- Project Folder
The Project Folder stores temporary work.
Examples:
campaign project
launch project
newsletter batch
course absorption block
client sprint
one-off report
ad test
funnel build
Project folders may contain temporary working files, drafts, exports, notes, and outputs.
Project folders must not become permanent context unless files are promoted deliberately.
Standard Folder Structure
The recommended base structure for MWMS context work is:
Context Area/
Context Library/
Skills/
Projects/
Archive/
For an offer or client, the recommended structure is:
Offer Or Client Name/
01 Raw Material/
02 IP Excavation/
03 Context Library/
04 Skills Or Employees/
05 Output Assets/
06 Audits/
07 Archive/
For internal MWMS Brain work, the recommended structure is:
Brain Name/
01 Source Material/
02 Canon Context/
03 Working Drafts/
04 Skills/
05 Outputs/
06 Audits/
07 Archive/
These structures may be simplified for low-risk work.
They should not be simplified for future AIBS client systems.
File Authority Rules
Rule 1: One Source Of Truth
Each active context file must have one approved source-of-truth location.
Do not maintain multiple live copies of the same context file.
Duplicated context creates drift.
Rule 2: Project Files Are Not Automatically Source Truth
A file created inside a project folder is not source of truth unless promoted.
Project work must be reviewed before it becomes library context.
Rule 3: Draft Files Are Not Active Context
Draft context files must not be treated as approved by AI Employees.
Use clear naming:
Draft
Under Review
Approved
Retired
Archived
Rule 4: Archive Is Not Active Context
Archived files are preserved for reference.
They should not be used by AI Employees unless the task specifically asks for historical comparison.
Rule 5: Retired Language Must Remain Visible
Retired Language files should remain easy for AI Employees to check.
Retired language prevents old claims, taglines, phrases, or positioning from creeping back into outputs.
Rule 6: Client Context Must Stay Separate
Client context must never be mixed with MWMS internal context or another client’s context.
Client context separation is mandatory.
Rule 7: Skills Must Reference Context, Not Duplicate It
A skill should read from the approved context library where possible.
Do not copy large context files into each skill.
The library stores what is true.
The skill defines what to do.
Naming Convention Rules
Files should use clear names that describe their role.
Good file names:
Right-Fit Client Profile
Offer Profile
Voice Architecture
Differentiation Profile
Objection Library
Methodology Map
Expert Thinking Rules
Customer Language Bank
Proof Library
Brand Visual Style
Retired Language
Compliance Notes
Bad file names:
Notes
Stuff
AI Things
Marketing
Copy Ideas
Final Final
New Doc
Random
MWMS file names should be readable by humans and AI Employees.
Version Naming
Use version markers when files are being refined.
Examples:
Voice Architecture v1
Voice Architecture v2
Offer Profile Draft
Offer Profile Approved
Sales Page Copy PRE-RESTYLE
Campaign Context May 2026
Do not use confusing version names.
Avoid:
final
final2
realfinal
latest maybe
new new
MWMS must keep version naming clear.
Multi-Offer Folder Patterns
MWMS recognizes three approved multi-offer patterns.
Pattern A: Shared Core With Offer Subfolders
Use when offers share the same brand, voice, and general methodology.
Structure:
Business Context Library/
Shared/
Offer A/
Offer B/
Shared may contain:
Voice Architecture
Brand Visual Style
Retired Language
Core Expert Thinking Rules
Brand-Level Differentiation
Offer folders may contain:
Right-Fit Client Profile
Offer Profile
Objection Library
Offer-Specific Differentiation
Proof Library
Campaign Context
Use this pattern when the offers are related but not identical.
Pattern B: Separate Libraries
Use when offers belong to separate brands, different voices, different markets, or different business models.
Structure:
Context Library – Brand A/
Context Library – Brand B/
This prevents context contamination.
Use this pattern when brand voice, buyer, or offer logic differs enough that shared context would confuse AI output.
Pattern C: One Library With Offer-Tagged Files
Use when offers have heavy overlap and only minor differences.
Structure:
Business Context Library/
Voice Architecture
Right-Fit Client Profile – Offer A
Right-Fit Client Profile – Offer B
Offer Profile – Offer A
Offer Profile – Offer B
Objection Library – Offer A
Objection Library – Offer B
Use carefully.
If AI Employees start mixing offers, upgrade to Pattern A or Pattern B.
Client Folder Standard
Future AIBS client systems require strict folder separation.
Recommended client structure:
Client Name/
Their Library/
Their Skills/
Working/
Outputs/
Audits/
Archive/
Their Library may include:
Client Voice Architecture
Client Offer Profile
Client Right-Fit Client Profile
Client Objection Library
Client Methodology Map
Client Expert Thinking Rules
Client Customer Language Bank
Client Proof Library
Client Compliance Notes
Their Skills may include:
Client Content Skill
Client Report Skill
Client Voice Checker Skill
Client Offer Copy Skill
Client Support Reply Skill
Working may include:
current drafts
campaign work
temporary analysis
in-progress outputs
Outputs may include:
approved deliverables
sent reports
published assets
final files
Audits may include:
context audit
skill audit
output quality review
client approval review
Archive may include:
completed projects
old versions
retired materials
Client Privacy Rules
Client context must be isolated.
MWMS must not:
mix client files with MWMS internal files
reuse one client’s language for another client
copy client stances into MWMS content
store client confidential information in general MWMS context
mount or load unrelated client context
allow client-specific skills to influence other client outputs
Client context is client-owned unless contract, approval, or governance says otherwise.
Project Folder Rules
Projects are temporary work containers.
Examples:
Q2 Launch
Newsletter May Batch
VEO3 Campaign Set
Client Onboarding Sprint
Offer Audit
Course Absorption Block
Ad Testing Sprint
Project folders may include:
briefs
drafts
temporary source files
exports
output versions
test files
review notes
Project folders must be cleared, archived, or promoted after the project ends.
Project Completion Outcomes
At the end of a project, each useful file must be classified.
Possible outcomes:
Promote To Context Library
Keep As Output Asset
Archive
Delete If Safe
Park For Later
Merge Into Existing Context
No project folder should remain messy forever.
Raw Material Folder Rules
Raw material folders store source material before it is processed.
Examples:
course files
client notes
emails
sales pages
voice memos
survey exports
customer interviews
screenshots
transcripts
Raw material is source evidence.
It is not automatically approved context.
Raw material must be extracted, reviewed, and transformed before becoming active context.
Output Asset Folder Rules
Output Asset folders store created deliverables.
Examples:
landing pages
emails
ad scripts
VEO3 scripts
lead magnets
webinar outlines
reports
sales pages
client deliverables
Output assets are not automatically context files.
If an output asset contains useful durable rules or language, extract that value into the context library rather than treating the whole output as context.
Archive Rules
Archive folders preserve old material without keeping it active.
Archive when:
project is complete
file is outdated
offer is retired
skill is retired
campaign is finished
old language should be preserved for reference
duplicate file is no longer active
client engagement is complete
Archive files should not be deleted unless deletion is approved.
Archive files should not be read by default for active output generation.
Retired Language Rules
Every active offer or brand library should have a Retired Language file when language changes over time.
Retired Language may include:
old taglines
old positioning
banned phrases
outdated claims
compliance-sensitive wording
language that no longer reflects the offer
phrases the founder dislikes
old campaign slogans
words that make copy sound generic
Retired Language is active context.
It exists to prevent old language from returning.
Update Trigger Rules
Context libraries must be updated when important changes occur.
Update triggers include:
new offer created
offer retired
offer repositioned
buyer changes
pricing changes
voice changes
methodology changes
new objections appear
new proof appears
compliance constraints change
customer language changes
AI output starts sounding wrong
old language resurfaces
skills require repeated correction
client feedback changes direction
campaign data reveals new learning
sales conversations reveal repeated friction
Do not let known context drift sit unresolved.
Context drift compounds.
Monthly Folder Check
For active libraries, MWMS should perform a light folder check monthly.
The monthly check asks:
Are files in the right folders?
Are there duplicates?
Are old project files still active?
Are draft files clearly marked?
Are retired files archived?
Is Retired Language visible?
Are client files isolated?
Are any files named poorly?
Are any folders becoming cluttered?
This check should be quick.
It prevents later cleanup pain.
Quarterly Library Audit
Each active context library should receive a deeper quarterly audit.
The quarterly audit asks:
Is each context file still true?
Is each context file still current?
Does the buyer profile still match the offer?
Does the offer profile still match reality?
Does voice still match current usage?
Are objections still current?
Is proof still approved?
Is old language creeping back?
Are skills reading the correct files?
Are there conflicting files?
Should the library be refreshed, split, merged, parked, or retired?
Quarterly library audits should align with AI Skill Audit cycles where possible.
Audit Outcomes
Each library audit must end with one of the following outcomes:
Current
Minor Refresh Required
Major Refresh Required
Split Library
Merge Library
Park Library
Retire Library
Archive Old Files
Human Review Required
No audit should end without a decision.
Context Promotion Rules
Files should be promoted into active context only after review.
Promotion requirements:
source material reviewed
content extracted
file structured
specificity checked
human-reviewed where required
destination clear
source-of-truth location defined
duplicates checked
status assigned
Do not promote rough notes directly into active context.
Context Demotion Rules
Files should be demoted when no longer reliable.
Demotion outcomes:
Move To Archive
Move To Retired
Park For Review
Replace With Updated Version
Merge Into Current File
Delete If Safe And Approved
Demotion should be deliberate, not accidental.
Common Failure Modes
MWMS must prevent:
folders multiplying without rules
duplicated context files
client files mixed with MWMS files
draft files used as approved context
project files becoming accidental memory
old language returning
archive files used as active source
skills carrying duplicated context
AI Employees reading the wrong offer folder
multiple Brains using different versions of the same offer
context libraries becoming bloated
no audit cadence
unclear file authority
unclear source-of-truth location
wrong client context loaded
Governance Role
HeadOffice owns the MWMS Context Library Governance And Folder Map Standard.
HeadOffice is responsible for:
defining folder structure
preventing context duplication
protecting source-of-truth discipline
governing archive rules
governing context promotion and demotion
requiring client context isolation
ensuring skills read from correct context
ensuring folder checks and audits occur
protecting MCR from messy context transfer
protecting future AIBS client systems
Individual Brains may maintain their own libraries, but they must align with this standard.
AI Business Systems Brain governs future client-library packaging.
Offer Brain governs offer-specific interpretation.
Content Brain governs content-use context.
Sales Brain governs sales-use context.
Creative Brain governs creative-use context.
Compliance Brain governs claims-risk context.
Relationship To Other MWMS Standards
This standard supports and must align with:
MWMS Document Structure Standard
MWMS AI Agent Memory And Context Framework
MWMS AI Agent Skill Library Framework
MWMS AI Skill Builder And Audit Protocol
MWMS Offer Context Library Standard
MWMS Client IP Excavation Framework
MWMS Course Absorption Operating Rule
MWMS Messy Input Normalization Framework
MWMS AI Output Validation Standard
MWMS Brain Routing Rule
MWMS MCR Promotion To Brain Protocol
MWMS Page Naming Standard
MWMS Architecture Registry
MWMS AI Employee Role Card Standard
MWMS AI Employee Capability Stack Framework
MWMS AI Tool Permission And Access Framework
Content Brain VOC Grounded AI Copy Framework
Research Brain Voice Of Customer Extraction Framework
AI Business Systems Brain Canon
This standard governs the folder and source-of-truth structure that makes those standards usable at scale.
Drift Protection
This standard protects MWMS from:
context sprawl
folder chaos
duplicate files
stale source truth
client context leakage
wrong offer context usage
draft context activation
archive context misuse
project file confusion
skill-context duplication
retired language returning
AI Employees reading the wrong files
manual re-explaining of the same context
MCR being polluted with ungoverned working material
future AIBS client systems becoming unsafe or confusing
Any context folder that does not have clear active, draft, archive, and source-of-truth separation should be treated as a drift risk.
Architectural Intent
The architectural intent of the MWMS Context Library Governance And Folder Map Standard is to make context durable, findable, governed, and safe to reuse.
MWMS will increasingly depend on AI Employees that need to know what to read before acting.
Those AI Employees cannot rely on scattered documents.
They need a clean folder architecture that answers:
Where is the source truth?
Which files are active?
Which files are draft?
Which files are archived?
Which files belong to this client?
Which files belong to this offer?
Which files should skills read?
Which files must not be used?
Which files need updating?
What has been retired?
What can be safely reused?
When MWMS can answer these questions consistently, AI work becomes more reliable, more scalable, easier to govern, and safer for future client systems.
Change Log
v1.0 — Initial Draft
Created the MWMS Context Library Governance And Folder Map Standard as the governance standard for organizing, separating, updating, archiving, and maintaining context libraries, skills folders, project folders, raw material folders, output folders, client libraries, retired language files, and future AIBS context structures.
This standard defines primary folder layers, file authority rules, naming conventions, multi-offer patterns, client folder standards, privacy rules, project folder rules, raw material rules, archive rules, retired language rules, update triggers, folder checks, quarterly audits, promotion and demotion rules, failure modes, governance role, drift protection, and architectural intent.
Change Impact Declaration
Pages Created:
MWMS Context Library Governance And Folder Map Standard
Pages Updated:
None
Pages Deprecated:
None
Registries Requiring Update:
MWMS Architecture Registry
HeadOffice Page Registry
AI Business Systems Brain Page Registry
Offer Brain Page Registry
Content Brain Page Registry
Canon Version Update Required:
No
Change Log Entry Required:
Yes
Employee Impact Check
Employees impacted:
HeadOffice Manager Employee
AI Manager
AI Employee Router
Context Library Builder
Client IP Excavator
Offer Strategist Employee
Content Planner Employee
Creative Strategist Employee
Sales Strategist Employee
AI Business Systems Architect Employee
Course Absorption Agent
Required behaviour updates:
AI Employees must not treat random folders, project files, or draft files as approved context.
AI Employees must respect one source-of-truth file locations and avoid duplicating context across skills or folders.
AI Employees must keep client context isolated from MWMS internal context and other client systems.
AI Employees must check Retired Language where relevant to avoid resurrecting outdated claims, phrases, positioning, or brand wording.
AI Employees must route useful project outputs through review before promoting them into active context libraries.
END MWMS CONTEXT LIBRARY GOVERNANCE AND FOLDER MAP STANDARD v1.0