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, Research 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 AI Context Pack Template Standard.
This standard establishes the required structure for packaging selected context before an AI Employee, Brain, workflow, skill, tool, or manual build begins important work.
MWMS must not rely on random memory, scattered notes, vague prompts, or full-library overload when assigning AI work.
An AI Context Pack gives the AI Employee the right working context for the current task.
The Context Pack does not replace the full context library.
It selects the relevant parts of the library, standards, source material, workflow state, constraints, and output requirements needed for one specific work unit.
This standard exists because AI work fails when it receives:
too little context
too much context
wrong context
stale context
unclear source truth
unclear task boundaries
unclear output expectations
unclear review requirements
unclear handoff destination
The AI Context Pack Template Standard gives MWMS a repeatable way to prepare AI Employees to do focused work without drifting.
Scope
This standard applies to all MWMS work where context must be packaged before action.
This includes:
AI Employee tasks
Brain Room requests
AI Manager tasks
AI Employee Router tasks
Course Absorption work
Newsletter Intelligence work
Offer evaluation work
Content Brain work
Creative Brain work
Sales Brain work
Conversion Brain work
Research Brain work
Affiliate Brain work
Ads Brain work
Client Brain work
future AIBS client systems
This standard applies before:
creating MCR pages
generating business assets
running skills
building lead magnets
building webinars
writing VEO3 scripts
drafting sales assets
drafting client reports
creating context libraries
reviewing outputs
preparing developer handoffs
routing cross-Brain work
This standard does not authorize development work, plugin changes, Supabase changes, WordPress changes, automation wiring, API implementation, or M developer action.
Core Definition
An AI Context Pack is a task-specific context bundle that tells an AI Employee what it needs to know before performing work.
A Context Pack may include:
current task
owning Brain
supporting Brains
source material
approved context files
relevant standards
workflow stage
output requirements
constraints
forbidden actions
risk level
human review requirement
handoff destination
missing context
assumptions
validation requirements
The Context Pack is the bridge between stored intelligence and active work.
Core Principle
The core principle of this standard is:
Use the smallest complete context pack needed for the task.
Too little context produces weak output.
Too much context produces confusion, cost, and drift.
Wrong context produces bad decisions.
A Context Pack should be focused, complete enough, and clearly tied to the current work unit.
Context Pack Use Cases
MWMS should use Context Packs for high-value or context-sensitive work.
Use Context Packs for:
MCR page creation
course absorption closeout
client Brain intake
offer evaluation
AI Employee task execution
content asset creation
sales asset creation
creative asset creation
lead magnet creation
webinar creation
developer handoff
cross-Brain requests
future client-facing AIBS work
Context Packs may be simplified for low-risk internal drafting.
Context Packs should not be skipped for high-risk, public-facing, client-facing, MCR, compliance, finance, paid traffic, or developer-related work.
Full Context Pack Template
Use the following structure for full context-sensitive work.
Task Name:
Current Request:
Work Type:
Owning Brain:
Supporting Brains:
Primary Source Material:
Approved Context Library:
Selected Context Files:
Relevant Standards:
Current Workflow Stage:
Required Output:
Output Destination:
Status Of Output:
Known Constraints:
Developer Boundary:
Client Boundary:
Tool Permission Boundary:
Risk Level:
Risk Type:
Human Review Required:
Relevant Past Decisions:
Do Not Use / Ignore:
Forbidden Actions:
Missing Context:
Assumptions:
Validation Requirement:
Handoff Destination:
Expected Outcome:
Next Action After Output:
Notes:
Field Definitions
Task Name
Defines the work being performed.
Example:
Create Context-Grounded Lead Magnet Funnel Page
Current Request
Captures the user’s current instruction.
Example:
User asked for next full page output from the Meera course absorption.
Work Type
Classifies the work.
Examples:
MCR Page Draft
Course Absorption
Context Library Build
Skill Build
Asset Build
Client Intake
Developer Handoff
Output Review
Owning Brain
Defines the Brain responsible for the work.
Examples:
HeadOffice Brain
Content Brain
Offer Brain
AI Business Systems Brain
Sales Brain
Supporting Brains
Defines Brains that may influence or review the work.
Examples:
Research Brain
Compliance Brain
Experimentation Brain
Creative Brain
Primary Source Material
Defines the source being used.
Examples:
course file
uploaded transcript
approved MCR page
client document
newsletter email
offer page
screenshot
Approved Context Library
Defines which context library applies.
Examples:
MWMS Offer Context Library
Client Context Library
Content Brain Canon
AI Business Systems Brain Canon
Selected Context Files
Lists the specific files being used.
Examples:
Offer Profile
Voice Architecture
Objection Library
Proof Library
Retired Language
Relevant Standards
Lists the MWMS standards that apply.
Examples:
Document Structure Standard
Course Absorption Operating Rule
AI Context Activation And Usage Protocol
AI Output Validation Standard
Current Workflow Stage
Defines where the task sits.
Examples:
Intake
Extraction
Drafting
Review
Closeout
Audit
Parked
Required Output
Defines what must be produced.
Examples:
full MCR page output
absorption report
skill record
context file
handoff brief
validation checklist
Output Destination
Defines where output should go.
Examples:
MCR
HeadOffice Review
Brain Room
Parking System
Client Review
M Developer Handoff
Status Of Output
Defines whether output is draft, active, review-only, or ready.
Examples:
Draft For MCR
Draft
Review Required
Manual Use
Approved
Known Constraints
Defines limits.
Examples:
do not cite course inside page output
do not create development tasks
do not touch M’s build
do not create duplicate page
Developer Boundary
Defines whether developer work is relevant.
Default:
No Development Action Authorized By This Page
Client Boundary
Defines whether client context is involved.
Examples:
Internal MWMS only
Client-specific
Client review required
Tool Permission Boundary
Defines tool limits.
Examples:
Provided input only
Read only
No external tools
No write action
Risk Level
Defines risk.
Examples:
Low
Operational
MCR Governance
Client
Compliance
Developer
Financial
Paid Traffic
Risk Type
Defines the risk area.
Examples:
duplicate page risk
source truth risk
client privacy risk
unsupported claim risk
wrong output format risk
Human Review Required
Defines whether human approval is needed.
Examples:
Yes
No
Required before MCR save
Required before client use
Relevant Past Decisions
Captures important approved decisions.
Examples:
wait until user says finished
course absorption must be selective
MCR is source of truth
Do Not Use / Ignore
Defines what should not influence the task.
Examples:
course hype
generic lesson summaries
weak material
stale memory
unapproved draft files
Forbidden Actions
Defines what the AI must not do.
Examples:
do not invent proof
do not create duplicate pages
do not imply development action
do not treat draft context as approved
Missing Context
Lists gaps.
Examples:
missing proof
missing client review
missing offer profile
missing current screenshot
Assumptions
Lists any assumptions that remain.
Assumptions should be minimized.
Validation Requirement
Defines how output should be checked.
Examples:
MCR structure validation
source grounding validation
Brain routing validation
context library validation
compliance validation
Handoff Destination
Defines where the work goes next.
Examples:
HeadOffice review
MCR entry
Brain registry update
human review
parked queue
Expected Outcome
Defines what useful result should happen.
Example:
New page strengthens MWMS context governance and can be added to MCR after review.
Next Action After Output
Defines the next step.
Examples:
create next page
update registry
perform closeout
park remaining material
review before MCR save
Short Context Pack Template
Use the short version for lower-risk work.
Task:
Owning Brain:
Source:
Context Needed:
Required Output:
Constraints:
Review Required:
Destination:
Next Action:
Context Pack Selection Rules
Rule 1: Start With The Task
Do not gather context before understanding the task.
Rule 2: Select Only Relevant Context
Do not load every file if only three files matter.
Rule 3: Current Request Controls The Session
The user’s current instruction must remain visible.
Rule 4: Approved Context Beats Memory
If approved context exists, use it instead of relying on memory.
Rule 5: Draft Context Must Be Marked
If context is draft, the output is draft.
Rule 6: Missing Context Must Be Flagged
Do not invent missing context.
Rule 7: Forbidden Actions Must Be Clear
The AI Employee must know what not to do.
Rule 8: Handoff Must Be Defined
The output must have a destination.
Rule 9: Risk Determines Review Level
Higher risk requires stronger review.
Rule 10: Context Pack Should Travel With Handoffs
If the task moves to another Brain or Employee, the relevant context must travel with it.
Context Pack Risk Levels
Low Risk
Examples:
internal brainstorming
light draft
non-public internal note
Review:
light review
Operational Risk
Examples:
workflow rule
internal process page
skill draft
Review:
human or HeadOffice review
MCR Governance Risk
Examples:
MCR page
canon candidate
standard
protocol
framework
Review:
MCR structure and registry review
Client Risk
Examples:
client report
client context library
client skill
Review:
client boundary and human review required
Compliance Risk
Examples:
ads
claims
testimonials
health, finance, income, affiliate content
Review:
Compliance Brain review required
Developer Risk
Examples:
code instructions
WordPress file edits
Supabase instructions
Make.com wiring
Review:
current evidence and M review required
Financial Risk
Examples:
budget recommendations
ROI assumptions
paid tools
investment decisions
Review:
human review required
Paid Traffic Risk
Examples:
Google Ads assets
YouTube ad scripts
campaign structure
claim-sensitive ad angles
Review:
Ads Brain and Compliance Brain review required
Context Pack Examples
Example 1: Course Absorption Page Creation
Task Name:
Create MWMS AI Context Pack Template Standard
Work Type:
MCR Page Draft
Owning Brain:
HeadOffice Brain
Supporting Brains:
AI Business Systems Brain, Course Absorption System, AI Manager
Primary Source Material:
Meera Kothand Your AI Brain Build course
Approved Context Library:
MWMS Course Absorption rules and MCR document standards
Relevant Standards:
Document Structure Standard
Course Absorption Operating Rule
AI Agent Memory And Context Framework
AI Agent Skill Library Framework
Required Output:
Full MCR page output
Known Constraints:
No citations inside page output
No development action authorized
Follow MWMS document structure
Human Review Required:
Required before MCR save
Handoff Destination:
MCR draft entry
Example 2: Client Report Draft
Task Name:
Draft Client Monthly AI System Report
Work Type:
Client Report Draft
Owning Brain:
AI Business Systems Brain
Supporting Brains:
HeadOffice Brain, Operations Brain, Data Brain
Primary Source Material:
Client workflow logs and approved client context library
Selected Context Files:
Client Offer Profile
Client Reporting Preferences
Client Approval Rules
Client Workflow Notes
Risk Level:
Client Risk
Human Review Required:
Client-facing review required
Forbidden Actions:
Do not invent performance data
Do not use another client’s context
Do not include unapproved recommendations
Example 3: Developer Handoff
Task Name:
Prepare M Developer Handoff For WordPress Plugin Fix
Work Type:
Developer Handoff
Owning Brain:
HeadOffice Brain
Supporting Brains:
Dev Console, AI Manager
Primary Source Material:
current screenshot and file content
Relevant Standards:
Developer Handoff Precision Skill
AI Agent Memory And Context Framework
Required Output:
exact developer brief
Risk Level:
Developer Risk
Human Review Required:
Yes
Forbidden Actions:
do not guess file paths
do not touch unrelated systems
do not assume current state from memory
Context Pack Failure Modes
MWMS must prevent:
starting work without a task boundary
using memory instead of source files
loading too much irrelevant context
loading too little context
forgetting current user instruction
forgetting output destination
forgetting human review requirement
missing developer boundary
missing client boundary
missing tool permission boundary
missing risk level
missing forbidden actions
allowing stale context to guide high-risk work
Context Pack Validation Checklist
Before using a Context Pack, check:
Is the task clear?
Is the owning Brain clear?
Are supporting Brains clear?
Is source material identified?
Are selected context files relevant?
Are applicable standards listed?
Is workflow stage known?
Is required output clear?
Is destination clear?
Are constraints clear?
Are forbidden actions clear?
Is risk level defined?
Is human review requirement defined?
Is missing context flagged?
Are assumptions marked?
Is validation requirement clear?
Is next action clear?
If several answers are missing, the Context Pack is not ready.
Governance Role
HeadOffice owns the MWMS AI Context Pack Template Standard.
HeadOffice is responsible for:
defining required Context Pack fields
ensuring high-risk work uses adequate context packs
preventing AI Employees from acting on incomplete context
ensuring output destination is defined
ensuring review requirements are included
ensuring context travels through handoffs
ensuring client context remains isolated
ensuring developer boundaries are included when relevant
Individual Brains may create specialized Context Pack variations, but they must align with this standard.
AI Business Systems Brain may create future client-specific Context Pack versions.
AI Manager and AI Employee Router may later operationalize context pack usage, but no technical build is authorized by this standard alone.
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 AI Skill Installation And Usage Protocol
MWMS Offer Context Library Standard
MWMS Context Library Governance And Folder Map Standard
MWMS AI Context Activation And Usage Protocol
MWMS Tool-Agnostic Context Portability Protocol
MWMS AI Brain Build Sequence Framework
MWMS AI Brain Readiness Review Checklist
MWMS AI Brain Audit And Decay Prevention Framework
MWMS Context-Driven Asset Builder Framework
MWMS AI Output Validation Standard
MWMS Brain Routing Rule
MWMS MCR Promotion To Brain Protocol
MWMS Page Naming Standard
MWMS Architecture Registry
AI Business Systems Brain Canon
This standard provides the reusable context package structure that supports active AI work.
Drift Protection
This standard protects MWMS from:
wrong context selection
missing task boundaries
AI Employees acting on stale memory
context overload
context underload
missing output destination
missing review gates
missing risk level
missing developer boundary
missing client boundary
handoff context loss
skills running without required context
future AIBS systems using incomplete context packs
Any high-value AI task started without a clear Context Pack should be treated as a drift risk.
Architectural Intent
The architectural intent of the MWMS AI Context Pack Template Standard is to make context selection repeatable, visible, and governable.
MWMS will increasingly depend on AI Employees performing specific work inside larger workflows.
Those AI Employees need the right context for the task, not unlimited memory and not empty prompts.
The long-term goal is that every important MWMS work unit can answer:
What is the task?
Which Brain owns it?
What source material applies?
Which context files are required?
Which standards govern it?
What output is required?
What must not happen?
What risk exists?
Who must review it?
Where does the output go?
What happens next?
When MWMS can answer these questions consistently, AI work becomes more focused, safer, easier to hand off, and easier to scale.
Change Log
v1.0 — Initial Draft
Created the MWMS AI Context Pack Template Standard as the standard for packaging selected task-specific context before AI Employees, Brains, skills, tools, or manual workflows perform important work.
This standard defines the Context Pack purpose, scope, definition, use cases, full template, field definitions, short template, selection rules, risk levels, examples, failure modes, validation checklist, governance role, drift protection, and architectural intent.
Change Impact Declaration
Pages Created:
MWMS AI Context Pack Template Standard
Pages Updated:
None
Pages Deprecated:
None
Registries Requiring Update:
MWMS Architecture Registry
HeadOffice Page Registry
AI Business Systems Brain Page Registry
AI Skill Registry
Canon Version Update Required:
No
Change Log Entry Required:
Yes
Employee Impact Check
Employees impacted:
HeadOffice Manager Employee
AI Manager
AI Employee Router
Brain Room Assistant
Course Absorption Agent
Context Library Builder
Client IP Excavator
Skill Auditor
Offer Strategist Employee
Content Planner Employee
Creative Strategist Employee
Sales Strategist Employee
Research Analyst Employee
AI Business Systems Architect Employee
Required behaviour updates:
AI Employees must use an appropriate Context Pack before performing high-value, high-risk, cross-Brain, MCR, client-facing, compliance-sensitive, paid traffic, finance, or developer-related work.
AI Employees must identify task, owning Brain, source material, selected context, relevant standards, required output, constraints, forbidden actions, risk level, human review requirement, handoff destination, and next action where relevant.
AI Employees must not rely on broad memory when an approved context library or task-specific context pack is required.
AI Employees must flag missing context, stale context, unclear destination, or review gaps before producing high-value outputs.
END MWMS AI CONTEXT PACK TEMPLATE STANDARD v1.0