CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ce-prototype

Build a throwaway prototype to answer how something should work, feel, or read. Use when committing the wrong answer would be expensive to unravel and a cheap sketch cannot settle it. Not a rough visual probe during brainstorming, not for deciding what to build, not polishing a feature that already works, not implementing the real thing.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

78%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 unusually disciplined instruction-only skill: every token is a rule, the phase sequence is gated with explicit authorization and stop conditions, and the progressive disclosure is exemplary — every reference is conditionally loaded and verified present. The residual costs are minor: a duplicated seeing/driving explanation, a few over-compressed sentences, and operational detail that lives entirely in references.

DimensionReasoningScore

Conciseness

The body is dense, imperative, and assumes Claude's competence with zero padding, but the seeing-vs-driving distinction is stated twice (intro paragraph and again under "Build it") and a few sentences are contorted enough to cost re-reading ("the offer to add that line comes before the root is resolved, or accepting it cannot help this run"). Not 5 because that repetition and tangling means not every token earns its place; well above 3 since nothing explains concepts Claude already knows.

4 / 5

Actionability

Concrete, executable specifics are present: exact run paths (`.context/compound-engineering/ce-prototype/<date>-<slug>/` with a `/tmp` fallback), a literal probe command (`git -C <repo root> check-ignore -q .context/compound-engineering/` with the trailing-slash caveat), and the named `decisions.md` capsule. Not 5 because much operational detail is delegated to references and some inline guidance stays abstract ("speak only when they can act on something new, in one short line naming what happened"); instruction-only framing keeps it at 4 rather than lower.

4 / 5

Workflow Clarity

A clear Scope → Build → Keep the decisions → Apply sequence with explicit gates ("Do not build until authorized", the no-human stop, the gitignore probe before root resolution) and an undo/recovery path for the risky overlay case ("restore only the files you changed. If you cannot undo them cleanly, name the files you left modified"). Not 5 because several checkpoints (go-ahead message, annotation-loop pacing) live in references rather than inline and some validation is implicit ("Update the capsule when you are confident").

4 / 5

Progressive Disclosure

The ~53-line body is a genuine overview: six reference files, all verified to exist, each with an explicit load condition — including when NOT to load one ("A question settled by driving does not load it") — plus bundled `scripts/light-webserver.js` and `assets/annotate.*` used via a SKILL_DIR anchor. References are one level deep, pointing only to siblings and bundled files, never to deeper content. This matches the clear-overview, well-signaled, one-level-deep anchor.

5 / 5

Total

17

/

20

Passed

Description

82%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 tightly written description with an explicit and concrete Use-when clause, strong boundary carve-outs against sibling skills, and the natural trigger term "prototype". The only real gap is specificity: it names a single action rather than the several concrete things the skill actually does, leaving a reader to infer scope from exclusions.

Suggestions

Name 1-2 more of the skill's concrete actions in the description (e.g., that it scopes the question, builds at the fidelity the question needs, and captures decisions for the next skill) so specificity does not rest on one verb plus exclusions.

Add one or two common synonyms for the core trigger (e.g., "mockup" or "spike") so users who do not say "prototype" still land on this skill.

DimensionReasoningScore

Specificity

"Build a throwaway prototype to answer how something should work, feel, or read" names one concrete action and its purpose, but the rest of the description is exclusion boundaries ("Not a rough visual probe... not polishing... not implementing the real thing") rather than an enumeration of the skill's concrete actions. Not 4 because it does not list several specific actions; not 2 because the action and its output are concrete, not generic.

3 / 5

Completeness

Explicitly answers what ("Build a throwaway prototype to answer how something should work, feel, or read") and when ("Use when committing the wrong answer would be expensive to unravel and a cheap sketch cannot settle it") plus explicit not-when clauses. Not 4 because the when-clause is concrete and trigger-like rather than a generic "when working with X".

5 / 5

Trigger Term Quality

"prototype", "throwaway", "sketch", and the natural phrasing "how something should work, feel, or read" give good keyword coverage a user would actually say. Not 5 because common synonyms like mockup, wireframe, or spike are missing ("sketch" appears only inside an exclusion clause).

4 / 5

Distinctiveness Conflict Risk

The description carves a clear niche with distinct triggers and explicitly excludes adjacent territories: "Not a rough visual probe during brainstorming, not for deciding what to build, not polishing a feature that already works, not implementing the real thing." Conflict risk with sibling skills is minimal.

5 / 5

Total

17

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

Total

15

/

16

Passed

Repository

EveryInc/compound-engineering-plugin

Path

skills/ce-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.