# zlh-grind

GPT Execution Workspace for ZeroLagHub

This repository exists to preserve execution continuity for long-running implementation work performed with GPT.

It is not a source-of-truth architecture repository.
Canonical documentation lives in:

jester/knowledge-base

---

## Assistant Prompt

You are údar.

In this repository you act as the execution assistant.

Your role is to:

- implement changes
- verify behavior
- document work
- maintain session continuity

Before doing work you must read:

SESSION_LOG.md
OPEN_THREADS.md
UPSTREAMS.md

Architecture decisions must not be made here.

If architectural interpretation is required:

Stop and defer to Claude or the canonical docs.

---

## What This Repo Is For

This repository tracks:

- GPT implementation work
- session handovers
- debugging notes
- execution reasoning
- cross-repo coordination

It serves as the execution ledger for the ZeroLagHub platform.

---

## What This Repo Is NOT

This repo must never become an architecture source.

Do not add:

- system diagrams
- architecture decisions
- platform policy
- runtime design

Those belong in:

jester/knowledge-base

---

## Platform Context (Current State)

ZeroLagHub consists of several core components.

### Control Plane

zlh-api

Responsibilities:

- container lifecycle orchestration
- database state
- agent provisioning payloads
- DNS / proxy coordination

---

### Execution Plane

zlh-agent

Responsibilities:

- container runtime provisioning
- game server install + management
- dev container provisioning
- runtime installs
- addon installs
- status reporting

---

### Artifact Distribution

zlh-artifacts

Artifact server hosts:

```
devcontainer/
  _catalog.json
  go/
  node/
  python/
  java/
  dotnet/

addons/
  code-server/
```

Artifacts are fetched by the agent at provisioning time.

Nothing provisioning-related is assumed to exist locally inside containers.

---

### Developer Containers

Dev containers support runtimes:

- node
- python
- go
- java
- dotnet

Runtime installs are:

- catalog driven
- artifact backed
- idempotent

Install root:

```
/opt/zlh/runtimes/<runtime>/<version>
```

Dev environment:

```
user: dev
home: /home/dev
workspace: /home/dev/workspace
```

---

## Repo Role in the Platform

This repository tracks execution work across all components:

- zlh-api
- zlh-agent
- zlh-portal
- zlh-artifacts
- infrastructure

It ensures GPT sessions remain consistent across days and chats.