Generic Template Package

“Take this. Use it. Build your own AI memory system.”

A complete, replicable template for implementing AI context management in any project — no Liana Banyan specific content.

What’s Included

GENERIC_TEMPLATE/
├── README.md                  ← Quick start guide
├── 00_READ_THIS_FIRST.md      ← Orientation (AI reads first)
├── 01_MASTER_CONTEXT.md       ← Source of truth template
├── 02_SYSTEM_REGISTRY.md      ← Components/systems template
├── 03_AI_CUSTOM_INSTRUCTIONS.md ← Platform instructions
├── 04_AGENT_SYNC.md           ← State tracking template
├── 05_HANDOFF_TEMPLATE.md     ← Session handoff format
├── 06_CONTEXT_PROTOCOL.md     ← Rules (generic, no edits needed)
├── 07_RECENT_CORRECTIONS.md   ← Error tracking template

All files contain [PLACEHOLDER] markers for your project-specific content.

Quick Start

Step 1: Copy the Template (5 minutes)

Copy GENERIC_TEMPLATE/ folder to your project root
Rename to CONTEXT_MANAGEMENT/

Step 2: Fill in Master Context (30-60 minutes)

Open 01_MASTER_CONTEXT.md
Replace all [PLACEHOLDER] markers with your data:
- Project description
- Key numbers and metrics
- Core components/products
- Team/stakeholder info
- Legal entity details

Step 3: Fill in System Registry (30-60 minutes)

Open 02_SYSTEM_REGISTRY.md
Document your systems:
- Name and purpose
- Status (Development/Live/etc.)
- Dependencies
- Integration points

Step 4: Set Up AI Platform (15 minutes)

Open 03_AI_CUSTOM_INSTRUCTIONS.md
Copy the instructions for your platform
Customize the [PLACEHOLDER] sections
Paste into ChatGPT/Claude/Cursor settings
Upload your context files

Step 5: Test (10 minutes)

Start new conversation
Ask about your project
Verify AI uses correct information

Total time to implement: 2-3 hours

Template Features

Fill-in-the-Blank Format

Every customizable section uses [PLACEHOLDER] markers:

## [PROJECT NAME] — Master Context

### [Category 1: e.g., Financial Metrics]
| Metric | Value | Context |
|--------|-------|---------|
| [Metric name] | **[Value]** | [Explanation] |

Consistent Structure

All templates follow the same organizational pattern:

• Clear section headers
• Tables for structured data
• Code blocks for diagrams
• Numbered procedures for protocols

Platform-Agnostic

Works with:

• ChatGPT Projects
• Claude Projects
• Cursor IDE
• Any LLM with file access

Agent Roles (Optional)

If you want multi-agent coordination:

• Define your own agent names
• Use the DROPZONE pattern
• Adapt the sync file structure

Customization Guide

Required Changes

No Changes Needed

Best Practices

Master Context

• Keep it factual — no speculation
• Update rarely — only verified changes
• Be precise — exact numbers, not estimates

System Registry

• Document dependencies explicitly
• Note integration points
• Track status for each component

Agent Sync

• Update after every significant accomplishment
• Be specific about what was done
• Include file paths for outputs

Corrections Log

• Check before generating content
• Log immediately when found
• Search for duplicate errors

Scaling Up

When to Keep File-Based

• < 100 active documents
• Single user
• Documentation mostly stable
• Cost-conscious

When to Add RAG

• 100+ active documents
• Multiple team members
• Frequent documentation changes
• Need cross-document synthesis

The template structure is RAG-ready. When you migrate, your markdown files chunk and embed without rewrite.

Support

The template is self-documenting. Each file explains its purpose and usage.

For Liana Banyan members:

• Setup guidance available
• Template customization assistance
• Integration troubleshooting

License

This template is provided freely. Use it for any project.

Attribution appreciated: “AI Context Management template from Liana Banyan”

Take the template. Build your memory. Never repeat yourself to AI again.

FOR THE KEEP!