CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

prototype

Asks a few questions, writes a one-page brief, builds a playable browser game as one HTML folder it plays and screenshots; with an engine set, a throwaway engine or paper prototype. Ends on proceed, pivot or kill. Use when the user says make me a game, wants to try a game idea or mechanic, needs a quick spike, asks to change a game it made, or says they love or hate it.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exemplary orchestrator body: lean routing logic, hard global constraints, and a one-level-deep reference structure where every cited file exists. Actionability and workflow clarity sit just below the top because execution and validation details are delegated to bundle and project files rather than inlined. The two 4-scoring dimensions could be nudged toward 5 by inlining one or two concrete validation checkpoints (e.g., the self-check pass condition) directly in the One-shot 2 section.

Suggestions

Inline the self-check's pass/fail condition (or a one-line 'build is done only when self-check passes' gate) in One-shot 2 so the validation checkpoint is explicit in the body, not only in references/one-shot-2c-self-check.md.

Add a one-line example of what a route-table input looks like in practice (e.g., a sample project.yaml engine line) so the routing decision is fully executable without opening external project files.

DimensionReasoningScore

Conciseness

The body is pure routing and project-specific rules — no explanation of concepts Claude already knows — packing dense, earned tokens like "3 PIVOT verdicts on the same concept force a KILL decision" and "Total effort is capped at 1 day". Every section delegates detail to a reference file rather than padding the overview. Not 4 because there is no over-explanation to trim.

5 / 5

Actionability

Concrete, executable directives throughout: the route table keyed on "Read `engine.name` from `project.yaml`" with explicit `[TO BE CONFIGURED]` handling, "Read `references/one-shot-1-card.md` ... and follow it", and hard rules like "If stuck after 2 hours of engine iteration, reframe the question or switch paths". Not 5 because actual execution detail (e.g., `.claude/genres/chipsets.yaml`, the card and build steps) is deferred to files whose content is not shown, leaving minor gaps an executor cannot resolve from the body alone.

4 / 5

Workflow Clarity

The sequence is clear and ordered: route selection ("Ask no question to choose the route"), then One-shot 0-5 or Phases 1-9, ending on the PROCEED/PIVOT/KILL verdict checkpoint, with "run and looked at before it is handed over" and a self-check file gating the build. Not 5 because most validation specifics live in the referenced files rather than being stated as explicit checkpoints in the body; well above 3 since checkpoints are present and signaled.

4 / 5

Progressive Disclosure

The body is a clean overview router: all 18 cited `references/*.md` files exist in the bundle, each signaled with "Read X now and follow it", one level deep with no nested references, and even file absence is handled ("With that file not there, growing is switched off"). This matches the anchor-5 structure of well-signaled one-level-deep references with appropriately split content.

5 / 5

Total

18

/

20

Passed

Description

83%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description with an explicit "Use when..." clause carrying natural trigger phrases and a concrete, multi-action statement of what the skill does. Both what and when are answered explicitly, matching the top completeness anchor. Minor room to grow: enumerate the engine/paper route's actions and add synonyms like "prototype" or "playtest" to the trigger list.

DimensionReasoningScore

Specificity

Lists several concrete actions — "Asks a few questions, writes a one-page brief, builds a playable browser game as one HTML folder it plays and screenshots" — plus the verdict outcome "Ends on proceed, pivot or kill". Not 5 because the engine route is summarized only as "a throwaway engine or paper prototype" without enumerating its actions; well above 3 since far more than 1-2 actions are named.

4 / 5

Completeness

Explicitly answers both: what ("Asks a few questions, writes a one-page brief, builds a playable browser game... it plays and screenshots") and when ("Use when the user says make me a game, wants to try a game idea or mechanic, needs a quick spike...") with concrete trigger phrases — a direct match for the anchor-5 example structure.

5 / 5

Trigger Term Quality

Natural user phrasings like "make me a game", "wants to try a game idea or mechanic", "needs a quick spike", and "says they love or hate it" are exactly what a user would say. Not 5 because common synonyms such as "prototype" or "playtest" are absent from the trigger list; clearly above 3 given the breadth of natural phrases.

4 / 5

Distinctiveness Conflict Risk

The game-prototyping niche is clear and triggers are game-specific, so conflict risk is low. Not 5 because "needs a quick spike" and "asks to change a game it made" could overlap with general technical-spike or iteration workflows; not 3 since the triggers are far more distinct than "somewhat specific".

4 / 5

Total

17

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

Donchitos/Claude-Code-Game-Studios

Path

.claude/skills/prototype/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.