CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ponytail

Apply Ponytail's reuse-first ladder when the user invokes /ponytail, asks for the simplest implementation, or wants to avoid over-engineering and new dependencies in a coding task. Preserve all requested behavior and required checks. Not a default for every coding task, a prose-compression mode, or a replacement for the repository's debugging, testing, or refactoring workflow.

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

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.

A lean, well-sequenced instruction-only skill: it defines a concrete reuse ladder with an explicit stopping rule, guards the known failure modes (partial delivery, false reuse, over-compression), and points to two real, well-organized reference files. It falls just short of top marks because there is no worked example and the rung-fallback behavior is implied rather than stated as an explicit loop.

Suggestions

Add a compact worked example (a sample request walked down the ladder to the chosen rung, plus the resulting report) to make the guidance copy-paste ready.

State the fallback explicitly: when a candidate fails the accessibility/data/concurrency/lifetime check, reject it and evaluate the next rung.

Trim or briefly gloss context-dependent cross-references such as 'Ponytail does not activate Caveman', which a reader without outside knowledge cannot interpret.

DimensionReasoningScore

Conciseness

The body is dense, skill-specific decision guidance with no explanation of concepts Claude already knows, but a few tokens could be trimmed — e.g. 'Ponytail does not activate Caveman' is opaque without outside context, and the savings-hypothesis closing could be one sentence shorter.

4 / 5

Actionability

Concrete, executable guidance throughout: an ordered 6-rung ladder with an explicit stopping rule ('Stop at the first option that satisfies the acceptance criteria'), enumerated check dimensions ('accessibility, data, concurrency, lifetime, and product behavior'), and concrete candidate examples (native date input, database constraint, standard cache). A worked example walking one request down the ladder would make it fully copy-paste ready, which keeps it below the top anchor.

4 / 5

Workflow Clarity

The sequence — establish behavior, choose an implementation, preserve quality and scope, report — is clear with real checkpoints (the ladder stopping rule, 'Read the candidate's behavior before reusing it', reporting 'what was verified') and a scope-preservation checklist. The reject-and-fall-through loop between rungs is implied rather than explicitly stated, so it sits at anchor 4 rather than 5.

4 / 5

Progressive Disclosure

A concise ~50-line overview with two clearly signaled, one-level-deep references ([provenance](references/provenance.md) and [the evaluation plan](references/eval-plan.md)), both of which exist in the bundle, with the detailed material appropriately split out and no nested references.

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 strong description that explicitly states what the skill does, when to activate it via three concrete triggers, and when not to activate it, yielding excellent completeness and distinctiveness. The main weakness is specificity: the capability rests on the named 'reuse-first ladder' concept without indicating what the ladder involves.

Suggestions

Enumerate one or two ladder rungs in the description (e.g. 'reuse existing helpers, the standard library, or installed dependencies before writing new code') so the core capability is concrete on its face.

Add natural synonyms such as 'keep it simple', 'minimal implementation', or 'don't overcomplicate' to the trigger clause to broaden keyword coverage.

DimensionReasoningScore

Specificity

It names the domain and two concrete actions — 'Apply Ponytail's reuse-first ladder' and 'Preserve all requested behavior and required checks' — but the ladder itself is never unpacked (no mention of reusing helpers, the standard library, or installed dependencies), so coverage is not comprehensive. It is below anchor 4 because it does not list several specific actions, and above anchor 2 because the actions it does give are concrete; voice is consistently third person, so no person penalty applies.

3 / 5

Completeness

It explicitly answers both questions: what ('Apply Ponytail's reuse-first ladder... Preserve all requested behavior and required checks') and when (three concrete trigger conditions), plus explicit negative scope ('Not a default for every coding task, a prose-compression mode, or a replacement for the repository's debugging, testing, or refactoring workflow'). Anchor 4's caveat that the 'when' could be more explicit does not apply here.

5 / 5

Trigger Term Quality

Natural phrases users would actually say are present — 'invokes /ponytail', 'asks for the simplest implementation', 'wants to avoid over-engineering and new dependencies' — but common synonyms such as 'keep it simple', 'minimal', or 'don't overcomplicate' are missing, keeping it below the comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

A clear niche (reuse-first minimal implementation) with a unique trigger (/ponytail) and explicit exclusions against the most likely conflicts — debugging, testing, refactoring, and prose-compression workflows — giving minimal conflict risk; only a faint overlap remains with generic 'keep it simple' requests, which the 'Not a default for every coding task' clause already narrows.

5 / 5

Total

17

/

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.

Repository

saski/arnesto

Path

.agents/skills/ponytail/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.