agentsclimarketplace

Gsp build strategy

Skill mike007jd/game-superpowers/publish/clawhub/gsp-build-strategy

Skills-first game development superpowers for Claude Code and Codex: build, audit, polish, and productionize game projects with reusable Agent Skills.

Install
npx -y skills add mike007jd/game-superpowers --skill gsp-build-strategy

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when deciding development mode, quality target, task granularity, or refactor policy for game work.

SKILL.md

2.5 KB, 538 tokens by cl100k_base, as published. Nobody here has run it

Game Build Strategy

Goal

Choose the right build strategy for the project.

Outputs

Follow the gsp-orchestrator output strategy:

  • inline (default): present build strategy and quality target in conversation.
  • minimal or full: write docs/game-studio/build-strategy.md and docs/game-studio/quality-target.md.

Use:

  • ./shared/reference/development-modes.md
  • ./shared/reference/quality-targets.md
  • ./shared/templates/build-strategy.md
  • ./shared/templates/quality-target.md

Development modes

  • yolo-super
  • guided-build
  • refactor-open
  • surgical-live

Quality targets

  • first-playable
  • polished-prototype
  • production-feature
  • live-patch

Selection rules

  • greenfield + narrow mechanic spike -> yolo-super + first-playable
  • greenfield + serious showcase build -> usually yolo-super or guided-build + polished-prototype
  • existing product-facing feature work -> usually guided-build or refactor-open + production-feature
  • shipped or live-risky -> surgical-live + live-patch

Planning policy

Match task size to the mode:

  • aggressive modes can use large coherent implementation chunks
  • production-feature work can use medium coherent chunks
  • live work should use smaller changes with tighter verification

Choose the exploration budget explicitly:

  • minimal when the task is narrow and already shape-locked
  • standard for normal product work
  • high for single-prompt showcase generation, benchmark runs, or any task where first-result quality matters more than token thrift

For high exploration budget work:

  • spend more tokens up front on concept lock, UX shape, and visual anchor decisions
  • prefer polished-prototype over first-playable unless the user explicitly wants only a spike
  • add runtime verification and screenshot critique before claiming completion
  • if the host supports subagents, default to builder + reviewer + verifier instead of a single uninterrupted build pass
  • if the task splits cleanly, allow multiple builders in parallel with a shared reviewer / verifier gate

Guardrail

Do not let “safe” planning ruin the result on low-risk greenfield work. Do not let “fast” planning justify broad rewrites on live products. For benchmark or showcase builds, do not optimize for token thrift if that weakens the result.

Gives 0 of the 12 instructions most roadmap strategy skills give in 538 tokens

Counted across 591 of the 672 authors here whose files we hold, read 2026-08-06

  • read product marketing context before asking questionsin 21 of 591, across 10 files
  • base price on perceived value, not costin 15 of 591, across 4 files
  • compact after finalizing a planin 14 of 591, across 9 files
  • differentiate tiers using features, limits, or supportin 14 of 591, across 3 files
  • use Van Westendorp to find acceptable price rangein 13 of 591, across 2 files
  • use MaxDiff to identify highly valued featuresin 13 of 591, across 2 files
  • map topics to buyer journey stagesin 12 of 591, across 6 files
  • Extract domain capabilities and classify subdomainsin 11 of 591, across 1 file
  • Define bounded contexts around consistency and ownershipin 11 of 591, across 1 file
  • Establish a ubiquitous language glossary and anti-termsin 11 of 591, across 1 file
  • Capture context boundaries in ADRs before implementationin 11 of 591, across 1 file
  • Open the strategic design template if neededin 11 of 591, across 1 file

Said here and by no other author read

  • match task size to mode
  • choose exploration budget explicitly
  • present strategy inline by default
  • allow parallel builders with shared reviewer gate
  • spend more tokens up front on high budget work
  • prefer polished prototype for high budget work

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.