# DeepSeek Harness Game Studios

![License](https://img.shields.io/badge/license-MIT-blue.svg)
![Version](https://img.shields.io/badge/version-0.1.6-orange.svg)
![Agents](https://img.shields.io/badge/agents-49-purple.svg)
![Skills](https://img.shields.io/badge/skills-76-green.svg)
![Hooks](https://img.shields.io/badge/hooks-14-orange.svg)
![Rules](https://img.shields.io/badge/rules-13-blue.svg)
![Port of CCGS](https://img.shields.io/badge/port%20of-CCGS%20v1.1.2-lightgrey.svg)

[![English](https://img.shields.io/badge/lang-English-blue.svg)](README.md) [![简体中文](https://img.shields.io/badge/lang-%E7%AE%80%E4%BD%93%E4%B8%AD%E6%96%87-red.svg)](README.zh-CN.md)

[Install](#install) · [Implementation status](#implementation-status) · [Project structure](#project-structure) · [Testing your customizations](#testing-your-own-customizations) · [Verification](#verification) · [License](#license)

Turn a single DeepSeek Harness session into a full game development studio.
**49 agents. 76 skills. One coordinated AI team.**

This is a **port of [Claude Code Game Studios](https://github.com/Donchitos/Claude-Code-Game-Studios) (CCGS)**
to DeepSeek Harness (DSH) — not a rewrite from scratch: the skills, agents, rules and docs come from
upstream (MIT); this repository provides the DSH-native runtime (Node hooks, skill generation,
machine-level install, plugin bridge).

## Provenance

| | |
|---|---|
| Upstream project | [`Donchitos/Claude-Code-Game-Studios`](https://github.com/Donchitos/Claude-Code-Game-Studios) |
| Upstream version this port tracks | **v1.1.2** (commit `b21fa0f`, 2026-09-29) |
| Upstream license | **MIT** (Copyright © 2026 Donchitos) — reused here |
| This repository | `deepseek-harness-game-studios` (DSGS) **v0.1.4** |
| Port strategy | Dual stack, one source: `.claude/` is the content source (upstream text, kept verbatim); `.dsh/**` is generated by `tools/build-dsh.mjs` |

Upstream targets Claude Code (`.claude/`, `.sh` hooks, `bash` one-liners, `@path` imports). None of that
runs under DSH (no `bash` tool, piped child processes are denied by the sandbox, `@path` is not expanded),
so instead of editing upstream content we generate a DSH-runnable equivalent and prove **no content loss**.

> This repository **is the plugin package**. The framework it installs is bundled under `payload/` — you never need to look at it.

## Install

### ① GUI

`Plugins` → `Add plugin` → paste `https://github.com/HYClub/deepseek-harness-game-studios` → **Install** → **restart `dsh web`**.

### ② Command line

```pwsh
dsh plugin --profile web add https://github.com/HYClub/deepseek-harness-game-studios
```

Then **restart `dsh web`**.

On first load the plugin installs the framework at machine level (`~/.dsh/dsgs` plus user-level
bootstrap skills). After that:

```
new working directory → new session → type /start
```

`/start` checks whether the workspace is initialised; if not it initialises it first, then walks you
through onboarding.

## What the plugin writes to your machine (and how to undo it)

| Path | What |
|---|---|
| `~/.dsh/dsgs/**` | framework payload (skills, agents, rules, hooks, tools, docs, instruction files) |
| `~/.dsh/skills/dsgs-init/SKILL.md`, `~/.dsh/skills/start/SKILL.md` | user-level bootstrap skills |

It never touches a project directory and never overwrites project files. To undo:

```pwsh
Remove-Item -Recurse -Force "$env:USERPROFILE\.dsh\dsgs"
Remove-Item -Recurse -Force "$env:USERPROFILE\.dsh\skills\dsgs-init","$env:USERPROFILE\.dsh\skills\start"
# then uninstall the plugin and restart dsh web
```

## Implementation status

### ✅ Implemented (all measured)

| Item | Status | Notes |
|---|---|---|
| Skills | ✅ | all 74 upstream skills ported, each invocable as `/name` |
| Agent role cards | ✅ | 49, as `.dsh/agents/<name>.md` + delegation conventions |
| Rules | ✅ | 13, including an equivalent for upstream's path-scoped rule engine: a PostToolUse hook injects the matching rule text after each write/edit |
| Hooks | ✅ | 12 upstream scripts → 14 Node rewrites, bridged via `dsh-hooks-claude-code`; all 9 upstream events covered |
| Hooks actually run | ✅ | mount 10/10 · protocol 12/12 · live events 6/9 |
| Config resolution | ✅ | `project.yaml` single source; Node ports of `resolve_config` / `__call resolve_setting` |
| Helper scripts | ✅ | every upstream `.sh` has a Node port (`.dsh/scripts/*.mjs`) that runs inside the sandbox |
| Content fidelity | ✅ | 258 files / 42,895 lines: 96.0% verbatim, 3.1% documented rewrites, 0.5% structural, 0.4% fuzzy, **0 unexplained loss** |
| Machine-level install | ✅ | the plugin ships the payload and installs it on first load; `/start` initialises an empty workspace (measured: 928 items) |
| `/statusline` | ✅ | studio status line `stage · posture · epic > story · Sprint n/m · engine` (+13 unit tests) |
| Nested delegation | ✅ | `subagent.maxDepth: 2` — parent → child → grandchild verified (DSH default is 1) |
| One-package install | ✅ | one package = hook bridge + status line projection + payload self-install |
| DSGS-only skills | ✅ | `statusline`, `dsgs-install` (not in upstream) |

### ❌ Not implemented / no equivalent (with reasons and substitutes)

| Item | Status | Why | Substitute |
|---|---|---|---|
| Docked status bar above the composer | ❌ | DSH's client roster only lists `@deepseek-ai/*` packages (measured: 60+ entries, ours absent even with a `dsh.client` declaration) | `/statusline` skill + `node .dsh/statusline.mjs`; the runtime half (turns/tokens/cache/context %) is already covered by DSH's own status bar |
| Pre-action reminders on PreToolUse | ❌ | `tools/pre-execute` only carries `allow/deny/cancel/ask`; `additionalContexts` is read from the final tool result; `agent.inject()` mid-execution has no effect | `validate-commit` / `validate-push` are also registered on **PostToolUse** (reminder arrives after the action); set the plugin row `advisory: deny` to block before the action |
| 3 hook events cannot be triggered from a prompt | ⚠️ | `Notification`, `PreCompact`, `PostCompact` have no DSH event source | mapping implemented and unit-tested, verified at mount level only (live coverage 6/9) |
| Per-agent tool allowlists / agent auto-discovery | ❌ | DSH's `subagent` does not accept a per-agent tool set and has no agent registry | role cards + `toolFilter` / `persona` conventions |
| `@path` imports | ❌ | DSH does not expand imports | `.dsh/docs/studio-rules.md` (merged), read on demand |
| `!`cmd`` pre-execution lines | ❌ | DSH has no pre-execution | rewritten as a "prerequisite command" block the model runs itself |
| Byte-for-byte bash ↔ Node parity | ⚠️ | `python3` in this machine's Git Bash is a Store placeholder | 2 named SKIPs, 1 case byte-identical |

## Project Structure

```
.claude/                     # Upstream content source (MIT, kept verbatim)
  settings.json              # Hooks, permissions, safety rules
  agents/                    # 49 agent definitions
  skills/                    # 74 skills (one directory per skill)
  hooks/                     # 14 scripts — 12 event hooks + yaml-helper.sh
  rules/                     # 13 path-scoped coding standards
  statusline.sh              # Upstream status line (kept as reference)
  docs/                      # workflow catalog, 39 templates, guidance
.dsh/                        # Generated by tools/build-dsh.mjs — do not edit by hand
  agents/  skills/  rules/  hooks/  scripts/  docs/
tools/                       # Generator, verifiers, Node ports, plugin, installers
dsh-src/                     # Hand-written DSH-only overlay (merged into .dsh/)
DSGS Skill Testing Framework/  # QA for the skills and agents themselves
design/ src/ assets/ tests/ prototypes/ production/   # Your project
```

## Testing your own customizations

DSGS is a template you are meant to edit. `DSGS Skill Testing Framework/` is how you check that an
edited or newly written skill still holds up — a catalog of all skills and agents, per-category quality
rubrics, behavioral specs, and templates for writing specs of your own.

```
/skill-test static [name]     # structural linter
/skill-test category [name]   # category-specific quality rubric
/skill-test spec [name]       # behavioral spec assertions
/skill-test audit             # coverage report across skills and agents
/skill-improve [name]         # test-fix-retest loop; keeps or reverts on score change
```

It tests the *framework*, never your game — game tests live in `tests/`.

## Verification

```pwsh
pwsh -File tools/verify-port.ps1        # 7 steps: generate → structure → refs → fidelity → hooks → mount → suites
node tools/check-dsh.mjs                # skills=76 agents=49 rules=13 hooks=14 errors=0 warnings=0
node tools/verify-fidelity.mjs          # unexplained 0 / no content loss (258 files, 42,895 lines)
node tools/verify-hooks-mount.mjs       # 10/10 mount cases
node tools/verify-hooks-bridge.mjs      # 12/12
node tools/verify-hooks-events.mjs      # 6/9 live + 3 mount-only
```

## Supporting this project (and upstream)

DSGS is a port; the original work — 74 skills, 49 agents, 13 rules and all the documentation — was
written by the **upstream author**, and their own support links live in the
[upstream repository](https://github.com/Donchitos/Claude-Code-Game-Studios). If this project saves you
time, please support **them** first. Issues and contributions about the DSH port itself belong in this
repository.

## License

MIT License — see [LICENSE](LICENSE). Upstream copyright © 2026 Donchitos; the port layer (generator,
Node hooks, plugin, installers, DSGS-only skills) is released under the same license.