@metaharness/gaming 0.1.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/LICENSE ADDED
@@ -0,0 +1,21 @@
1
+ MIT License
2
+
3
+ Copyright (c) 2026 rUv
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
package/README.md ADDED
@@ -0,0 +1,63 @@
1
+ # MetaHarness: gaming vertical
2
+
3
+ A ready-made multi-agent game-design pod scaffolded as a Claude Code harness. One command drops a working `claude-code` plugin directory on disk with four specialist agents wired up — a concept lead, a mechanics designer, a balance analyst, and a playtest critic — plus the project settings, slash commands, and MCP plumbing needed to actually run them. Aimed at solo designers, game-jam teams, and studios who want a structured pod for ideation and iteration instead of a single generalist prompt. This is a scaffold, not a game engine: it does not ship Unity/Unreal/Godot integrations, no asset pipeline, no live ops — it gives you the design loop, you bring the runtime.
4
+
5
+ ## Quickstart
6
+
7
+ ```bash
8
+ npx @metaharness/gaming@latest my-bot
9
+ cd my-bot && npm install && harness doctor
10
+ ```
11
+
12
+ The `harness doctor` step validates that Claude Code is on PATH, the plugin manifest parses, and each agent's model tier is reachable.
13
+
14
+ ## What you get
15
+
16
+ - `agents/concept-lead.md` — pitch and high-concept exploration (tier: **opus**)
17
+ - `agents/mechanics-designer.md` — core loop, systems, and rule specification (tier: **sonnet**)
18
+ - `agents/balance-analyst.md` — numeric tuning, curves, and economy modelling (tier: **sonnet**)
19
+ - `agents/playtest-critic.md` — adversarial critique, fun audit, edge-case hunting (tier: **haiku**)
20
+ - `commands/pitch.md`, `commands/loop.md`, `commands/balance.md`, `commands/playtest.md` — slash commands that route to the right agent
21
+ - `.claude/settings.json` — plugin manifest, permissions, and hook wiring
22
+ - `CLAUDE.md` — pod operating rules: who calls whom, when to fan out, when to converge
23
+
24
+ ## Advanced
25
+
26
+ ```bash
27
+ # Validate the scaffold and the model routing
28
+ harness doctor
29
+ # => OK plugin manifest
30
+ # => OK agents/concept-lead.md (claude-opus)
31
+ # => OK agents/mechanics-designer.md (claude-sonnet)
32
+ # => OK agents/balance-analyst.md (claude-sonnet)
33
+ # => OK agents/playtest-critic.md (claude-haiku)
34
+
35
+ # Lint frontmatter, agent tier hints, and command routing
36
+ harness validate
37
+ # => 4 agents, 4 commands, 0 errors
38
+
39
+ # Run the pod headless against a design brief
40
+ claude -p --plugin-dir ./my-bot "/pitch a co-op deckbuilder set on a generation ship"
41
+ # => concept-lead drafts pitch -> mechanics-designer proposes core loop
42
+ # => balance-analyst flags one runaway resource curve
43
+ # => playtest-critic returns 6 fun-risk findings
44
+ ```
45
+
46
+ ## FAQ
47
+
48
+ **Does this generate a playable game?**
49
+ No. It generates design artifacts — pitch docs, mechanic specs, balance tables, playtest critiques. Hand those to your engine of choice.
50
+
51
+ **Why four agents instead of one prompt?**
52
+ Because "design a game" is the kind of task where a single generalist regresses to genre cliches. Splitting concept / mechanics / balance / playtest forces each pass to defend its choices to the next agent.
53
+
54
+ **Can I swap the model tiers?**
55
+ Yes. Each agent's frontmatter has a `model` field. Drop `playtest-critic` to `haiku` for cheap iteration, bump `concept-lead` to `opus` when you want the high-concept work to stretch.
56
+
57
+ ## License
58
+
59
+ MIT. Built on metaharness (https://www.npmjs.com/package/metaharness).
60
+
61
+ ## Deep-dive
62
+
63
+ Full explainer gist: https://gist.github.com/ruvnet/89ab98e6e5f082a376170a941c9d1914
@@ -0,0 +1,37 @@
1
+ #!/usr/bin/env node
2
+ import { execSync } from 'node:child_process';
3
+ import process from 'node:process';
4
+
5
+ const [, , rawName, ...extra] = process.argv;
6
+ const name = rawName && !rawName.startsWith('-') ? rawName : 'my-bot';
7
+ const forwarded = rawName && rawName.startsWith('-') ? [rawName, ...extra] : extra;
8
+
9
+ const args = [
10
+ 'npx',
11
+ '--yes',
12
+ 'metaharness@latest',
13
+ name,
14
+ '--template',
15
+ 'vertical:gaming',
16
+ '--host',
17
+ 'claude-code',
18
+ '--force',
19
+ ...forwarded,
20
+ ];
21
+
22
+ try {
23
+ execSync(args.join(' '), { stdio: 'inherit' });
24
+ } catch (err) {
25
+ process.exit(typeof err?.status === 'number' ? err.status : 1);
26
+ }
27
+
28
+ process.stdout.write(
29
+ [
30
+ '',
31
+ ` Next steps:`,
32
+ ` cd ${name}`,
33
+ ` npm install && harness doctor`,
34
+ ` claude -p --plugin-dir . "/pitch a co-op deckbuilder set on a generation ship"`,
35
+ '',
36
+ ].join('\n'),
37
+ );
package/package.json ADDED
@@ -0,0 +1,41 @@
1
+ {
2
+ "name": "@metaharness/gaming",
3
+ "version": "0.1.0",
4
+ "description": "MetaHarness example — ready-made game-design pod (concept, mechanics, balance, playtest) for Claude Code (npx @metaharness/gaming my-bot)",
5
+ "homepage": "https://github.com/ruvnet/agent-harness-generator/tree/main/examples-packages/gaming",
6
+ "repository": {
7
+ "type": "git",
8
+ "url": "https://github.com/ruvnet/agent-harness-generator.git",
9
+ "directory": "examples-packages/gaming"
10
+ },
11
+ "license": "MIT",
12
+ "author": "rUv <ruv@ruv.net>",
13
+ "type": "module",
14
+ "bin": "./bin/scaffold.mjs",
15
+ "files": [
16
+ "bin/**",
17
+ "README.md",
18
+ "LICENSE"
19
+ ],
20
+ "scripts": {
21
+ "smoke": "node --check bin/scaffold.mjs"
22
+ },
23
+ "keywords": [
24
+ "metaharness",
25
+ "agent-harness",
26
+ "agent-harness-generator",
27
+ "gaming",
28
+ "vertical",
29
+ "claude-code",
30
+ "ai-agent",
31
+ "mcp",
32
+ "npx",
33
+ "scaffold"
34
+ ],
35
+ "publishConfig": {
36
+ "access": "public"
37
+ },
38
+ "engines": {
39
+ "node": ">=20.0.0"
40
+ }
41
+ }