prototype

Concept prototype before GDDs — throwaway HTML, Engine or Paper build, PROCEED/PIVOT/KILL/NOT ASSESSED. After /brainstorm and /setup-engine.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/prototype/SKILL.md

Quality

Content

77%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 exceptionally actionable, well-sequenced workflow skill with strong validation checkpoints and feedback loops throughout. Its main weaknesses are a padded, monolithic body: duplicated playtest guidance and inlined tool/reference catalogs that inflate token cost without adding navigational structure.

Suggestions

Deduplicate playtest-observation guidance ("Confusion is data", "What was confusing?", watch-silently protocol) — it appears in both the Paper path section and Phase 5; state it once and reference it.

Move the tool and channel catalogs (Love2D, Twine/Ink/Yarn Spinner, itch.io/Reddit/Discord distribution options) into a references/ file, keeping one-line pointers in SKILL.md.

Compress explanations of techniques Claude already knows (Wizard of Oz, think-aloud) to a one-line "when to apply" mention, cutting the multi-paragraph tutorials.

Dimension	Reasoning	Score
Conciseness	The bulk is non-obvious domain guidance (path selection by genre, sunk-cost rules, verdict routing), but padding is noticeable: playtest-observation advice ("Confusion is data", "What was confusing?") appears in both the Paper path and Phase 5, distribution channels (itch.io/Reddit/Discord) are described in the HTML path and re-pointed in Phase 5, and known techniques (Wizard of Oz, think-aloud protocol) are explained at length. Mostly efficient with several sections that could be tightened — anchor 3.	3 / 5
Actionability	Fully executable throughout: exact AskUserQuestion prompts and options, exact file paths (prototypes/[concept-name]-concept/REPORT.md, production/session-state/active.md), per-language comment syntax ("# in GDScript... -- in Lua"), hard caps, the GRAVEYARD entry template, and verbatim scripts for the engine multi-turn loop. Copy-paste ready and covers the common cases.	5 / 5
Workflow Clarity	Phases 1–9 plus Spike Mode are clearly sequenced with explicit validation checkpoints (sunk-cost rule, "If nobody can play it... NOT ASSESSED", vague-answer follow-ups), feedback loops for error recovery (engine iterate-until-playable loop, CD-gate APPROVE/CONCERNS/REJECT handling), and checklists (the KILL-confirmation boxes).	5 / 5
Progressive Disclosure	No references/ bundle exists, and external doc references (.claude/docs/templates/prototype-report.md, automation-modes.md, director-gates.md) are clearly signaled and one level deep. However, the ~630-line body inlines reference-type material — the genre→path table, Love2D and Twine/Ink/Yarn tool catalogs, and distribution-channel lists — that belongs in separate reference files. Anchor 3: content that should be separate is inline.	3 / 5
	Total	16 / 20 Passed

Description

75%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 dense, domain-specific description that clearly states what the skill produces and when it runs in the pipeline. Its weaknesses are phrasing "when" as workflow position rather than user intent, and missing natural trigger variations such as "playtest", "validate", or "spike".

Suggestions

Add an explicit user-intent trigger clause, e.g. "Use when the user wants to test whether a game concept is fun before writing GDDs, or wants a quick mid-production spike."

Include natural trigger synonyms users would actually say — "playtest", "validate the idea", "test the core mechanic", "spike" — alongside the existing verdict jargon.

Dimension	Reasoning	Score
Specificity	"throwaway HTML, Engine or Paper build, PROCEED/PIVOT/KILL/NOT ASSESSED" names concrete build paths and output verdicts, giving several distinct specifics. It falls short of 5 because actions are noun-listed ("prototype", "build") rather than a comprehensive enumeration of concrete operations.	4 / 5
Completeness	The "what" is clear (concept prototype as a throwaway build across three paths producing a verdict) and the "when" is explicitly stated via "before GDDs" and "After /brainstorm and /setup-engine", so it is not capped at 3. It falls short of 5 because the "when" is phrased as pipeline sequencing rather than concrete user-intent trigger phrases.	4 / 5
Trigger Term Quality	"Concept prototype", "throwaway", "HTML", "Engine", "Paper" are terms a user would naturally say. "PROCEED/PIVOT/KILL" is internal verdict jargon rather than a user trigger, and natural phrases like "playtest", "validate", "test if it's fun", or "spike" are missing — good coverage with a few gaps, matching anchor 4.	4 / 5
Distinctiveness Conflict Risk	Distinctive markers ("GDDs", "PROCEED/PIVOT/KILL", "/brainstorm", "/setup-engine") carve out a clear niche, but generic wording like "prototype" and "throwaway HTML" leaves minor overlap risk with adjacent skills. Mostly distinct with minor overlap risk — anchor 4, not the minimal-conflict profile of 5.	4 / 5
	Total	16 / 20 Passed

Validation

81%

Warnings & errors only

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

Validation — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (637 lines); consider splitting into references/ and linking	Warning
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	13 / 16 Passed

Repository: Donchitos/Claude-Code-Game-Studios
Path: .claude/skills/prototype/SKILL.md
Commit: b21fa0f

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.