Browse / Software development / ai-workspace-template
ai-workspace-template
hebert-ai
A conservative, repo-first template for AI-assisted multi-project workspaces.
Open hebert-ai/ai-workspace-template → On the boundary of the four rules.
agent-workflowsai-workspacedeveloper-toolsgithub-templatellmproject-managementtemplate
Works with any agent
ships AGENTS.md, the cross-tool instruction file.
In its own words
AI Workspace Template is a conservative, repo-first system for running an AI-assisted project workspace across many repos.
It gives a user one root workspace repo for planning, prioritization, shared agent guidance, templates, and workspace tooling, while keeping real project repos isolated under projects/.
This project is the canonical template product for that workspace model.
Why This Exists
Most multi-project AI workflows need one place to:
- track project status across many repos
- prioritize work across projects
- keep shared instructions for multiple LLM clients
- preserve personal workspace conventions across machines
- bootstrap new projects without turning the workspace root into a product repo
This template is designed around that split:
- root workspace repo for coordination
projects/for real project reposcustom/for user-owned agents, scripts, and status tracking- explicit sync for template-managed files only
Quick Summary
- Publish this repo as the canonical template source.
- Each user creates their own separate workspace repo from it.
- That user workspace repo becomes their persistent AI management workspace.
- Each real project lives under
projects/ /as its own repo. - Template updates are applied through explicit sync, never by overwriting
custom/orprojects/.
Product Model
This repo is the product repo for the template itself, not the day-to-day workspace a user operates from.
The intended deployment model is:
Quoting 24% of the README. Read the rest on GitHub ↗
Read from the repository's git tree on 2026-08-17.