TempleMap

TempleMap keeps your .md files up to date, has memory and generally makes working locally on non-coding tasks a TON more fun.

TempleMap is like NextJS for Markdown projects

Sick of setting up each folder you do AI knowledge work in? Thinking about where to put which .md file and how to keep them updated?

Installation

TempleMap is a Markdown repo template. There is no package install step.

git clone https://github.com/nikoatwork/TempleMap.git my-project
cd my-project

If this clone is becoming your own project, point origin at your project repo and keep TempleMap as the template remote:

git remote rename origin template
git remote add origin https://github.com/YOUR-USER/YOUR-REPO.git

Then open the folder with your agent of choice — Claude Code, Codex, Pi, or similar — and ask:

Read [[routines/onboard-workspace]] and guide me through onboarding.

The onboarding routine will help you adapt the template into a project-specific memory.

How it works

A file-based knowledge-work repo template.

The repo is the durable memory. Agents and humans work by updating canonical Markdown files, linking related notes, and keeping a compact history of what changed.

Core loop

source/input -> work item -> routine -> module/artifact -> consolidation -> review/check -> history

Main folders

• system/ — operating rules and repo conventions
• definitions/ — canonical shared concepts
• work/ — current task instances and execution state
• routines/ — reusable runbooks / entrypoints
• modules/ — reusable capabilities / domain logic
• artifacts/ — generated deliverables
• sources/ — raw evidence and inputs
• templates/ — expected file shapes
• checks/ — validation rules
• reviews/ — qualitative audits and critiques
• history/ — concise changelog and decision log

Use wiki-style links like [[definitions/example]] for internal references.