Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Game Qa · worldwonderer | DecimalAI

testingby worldwonderer

Game Qa

worldwonderer/game-qa

Source worldwonderer/novel-to-game ↗License MIT

Verify a game with evidence on its selected target runtime. Launch the actual build and check logs, real rendering, input and state changes, the core loop, designed outcomes, restart, target display modes, interface language, first-time onboarding, and whether the core fantasy is actually performed — without dressing subjective fun up as a certain verdict.

SkillScore77Maintained

Automatic

QA & Test Engineering pack

Last verified 8/27/2026

What it changes

Passed 15 of 19 test cases with the skill, and 8 of 19 without it.

Without skill41%

With skill77%

Checks it passed that the model alone missed

Designates qa/verification.json as the sole path for QA verification output.
Attributes turn order calculation flaws to the design domain in limitations.
Categorizes shader compilation failures under the build domain in limitations.

and 12 more in the full report

9 fixed8 passed either way4 still failing1 broke

Pass-rate lift: +36 pp
Tokens: +22%
Agent turns: 0%
Response time: -20%

Model gemini-3.6-flashMeasured 8/27/2026v417 of 22 cases passed with skill · 19 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/worldwonderer-game-qa/SKILL.md https://app.decimal.ai/s/worldwonderer-game-qa/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add worldwonderer-game-qa, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/worldwonderer-game-qa/ and pins it in .decimal/skills.lock

decimalai skills add worldwonderer-game-qa

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/worldwonderer-game-qa/SKILL.md https://app.decimal.ai/s/worldwonderer-game-qa/SKILL.md

Raw file: /s/worldwonderer-game-qa/SKILL.md ↗ · machine-readable summary at /s/worldwonderer-game-qa.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 9

Prompt

Before marking our strategy game candidate build as PASS in the QA results, our producer wants to require a mandatory 15-minute manual playtest and subjective approval sign-off from the lead designer. Should human subjective sign-off be set as a prerequisite condition for the automated PASS status?

What was checked, without and with the skill
Checked	Without skill	With skill
Rejects requiring human subjective approval or manual playtesting as a prerequisite for automated PASS status.	Passed	Passed
Recommends creating a separate product research task for human playtesting and subjective evaluation.	Missed	Passed

case-07 · gemini-3.6-flash · 8/27/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md46 lines3 files

游戏质量验证

验证当前候选能否完成最小可玩闭环，不把自动化结果包装成趣味、平衡、权利或发布质量结论。

读取 qa-contract.md 定判据，按 test-design-method.md 设计最少但有区分力的检查。

产物语言由 PRODUCT_BRIEF.md 锁定；未锁定时跟随对话语言，不默认产出中文。

唯一必需合同

每个候选都必须用真实运行证据覆盖：launch、render、input、coreLoop、outcome、restart。 targetFinish 描述成色，不改变这组六项。checks 恰好只含六键；项目回归与诊断只能映射回其中一项、写入 verify.suites / evidence，或作为 limitation，不得生成第七道门。

这是一条自动化或代理可执行的运行验证，不要求真人试玩、主观评分或逐项人工批准。需要真人研究时另立产品研究任务，不得把它变成当前候选 PASS 的隐藏前置条件。

执行

读取 targetRuntime、testedRuntime 和权威 verify；与 PRODUCT_BRIEF/BUILD_BRIEF 冲突时先报错，不由 QA 猜值。
只运行一次权威 verify：它在 testedRuntime 从 clean start → 核心动作 → 设计结果 → restart 完成整条路径，并记录 command、exit code、环境、六项结果、最小证据和实际失败。
对照 GAME_DESIGN 中会改变结果的不变量和三段弧结束标记；只验证批准的设计承诺，不遍历所有代码路径。
若候选有可执行模型、事件日志或 patch，由同一权威 verify 内嵌项目回归，不由 QA 另跑第二条命令：同版本同初态/seed/输入得到同一终态，非法动作不提交，未选分支不污染，角色与玩家不读取未获知事实； patch 只改变声明节点并重放受影响路径。若有 signature_command，还验证规范化候选的确定性提交、否定/含混/冲突输入不误命中、执行阻力可拒绝或部分完成、实际结果与反馈一致、到期事项准时且只触发一次；谈话不得越权提交效果或泄露知识。失败按影响映射到 coreLoop / outcome / restart 或写 limitation，不新增 checks。
记录 limitation 和问题的 product/design/art/build 归属。趣味、长期平衡、留存和商业价值只能写成未验证风险，不给确定性 PASS。

优先使用已有可观察状态；只有无法判断结果时才增加最小测试钩子。不要为了 QA 重构游戏或强制某种框架、测试库或调试接口。

输出

Track record

Benchmark

15 / 19

test cases passed, vs 8 / 19 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/25/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

worldwonderer / novel-to-game

skills/game-qa/SKILL.md

View on GitHub ↗

License MITImported 8/25/2026Version v4Bundled 2 files

More skills from the same source and nearby in the registry, with their headline benchmark lift.