CtrlK
BlogDocsLog inGet started
Tessl Logo

wagneripjr/prototype-spike

Turn a requirement or an unproven technique into a single self-contained HTML prototype that doubles as a design spike, rebuilt at high fidelity from the app's own source

97

1.72x
Quality

97%

Does it follow best practices?

Impact

100%

1.72x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

Overview
Quality
Evals
Security
Files

Quality

Content

92%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.

A sophisticated, well-engineered skill body: a gated multi-phase workflow with explicit validation, concrete and executable tooling guidance, and clean one-level-deep reference split. Its only weakness is length — a few expository framing passages could be trimmed without losing the methodology.

Suggestions

Tighten the rhetorical framing lines (e.g. the lab-notebook analogy, 'a disagreement about the prototype is a disagreement about the system') into terse procedural statements to lift conciseness toward the lean anchor.

Consider moving the Hard Rules scan-before-you-ship checklist into a reference file, keeping only the gate cross-references inline, to reduce body length while preserving the shipping gate.

A couple of the per-row LSP/fallback table rows repeat guidance already implied by the gates; collapse the most redundant rows to trim tokens.

DimensionReasoningScore

Conciseness

The body is information-dense and assumes Claude's competence — it never explains basics — but at ~230 lines it carries a few rhetorical framing passages ('Treat it the way you would a lab notebook') that, while load-bearing, could be tightened, so it sits just below the lean anchor.

4 / 5

Actionability

Highly executable: concrete tool invocations (ToolSearch select:LSP, mcp__claude-in-chrome__navigate), curl/grep commands, a copy-paste spike-report template, a harvest-table format, and a 9-item verification checklist cover the common cases.

5 / 5

Workflow Clarity

Phases are explicitly sequenced (0 ANCHOR → 5 CLOSE), each with a named gate (G0–G5) acting as validation checkpoints, plus feedback loops (re-validate, re-grep, delete dead CSS) and a checklist; the destructive secret/PII grep has validation, so no cap applies.

5 / 5

Progressive Disclosure

The body is an overview that pushes detail to seven well-signaled one-level-deep references (all real files in ./references/), each with a one-line description in the References section, making navigation easy.

5 / 5

Total

19

/

20

Passed

Description

100%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.

An exceptionally strong description: concrete and comprehensive in its actions, with explicit natural trigger phrases and clear negative boundaries that fence it off from neighbouring skills. It answers both what and when without resorting to vague or padded language.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'produces a single self-contained HTML prototype', 'Rebuilds existing screens at high fidelity from real source', 'derives one control per acceptance criterion', 'drives real data through a dev-server proxy', 'verifies every state in a real browser', 'reports what the spike settled' — with comprehensive coverage and no vague filler.

5 / 5

Completeness

Clearly answers both 'what' (the prototype/harvest/control/verification/report actions) and 'when' via a concrete 'Triggers on …' clause, satisfying the top anchor with explicit trigger phrases.

5 / 5

Trigger Term Quality

Explicit natural trigger phrases are given — 'prototype FR-NNN', 'clickable prototype of FR-NNN', 'make this requirement clickable', 'spike this technique' — covering the natural variations an engineer would actually say.

5 / 5

Distinctiveness Conflict Risk

A clear niche (requirement/technique spike) is reinforced by explicit 'NOT for …' boundaries distinguishing it from generative-design, production frontend, spec-playback, and ADR skills, yielding minimal conflict risk.

5 / 5

Total

20

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

Table of Contents