Content
40%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is token-efficient but is generic template text: no concrete formulas, code, or sequenced workflow for any of the named risk metrics, and it delegates all substance to a playbook file that is absent from the bundle. As shipped, the skill gives Claude nothing actionable to execute.
Suggestions
Replace the generic Instructions bullets with concrete guidance: key formulas or executable code for each named metric (e.g., historical vs. parametric VaR, CVaR from the tail of the loss distribution, Sharpe = (Rp - Rf)/sigma_p, Sortino using downside deviation, max drawdown from the equity curve).
Add a sequenced workflow with validation checkpoints, e.g.: 1) gather/normalize the return series, 2) compute each metric, 3) sanity-check outputs (e.g., CVaR >= VaR, Sortino >= Sharpe for the same series) before reporting.
Ship the referenced resources/implementation-playbook.md or inline its essential content — the body's only reference is a dangling pointer, so progressive disclosure currently leads nowhere.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and assumes Claude's competence with no concept explanations, but the "Comprehensive risk measurement toolkit..." intro and "Use this skill when" list duplicate the frontmatter description, and "The task is unrelated to risk metrics calculation" is vacuous filler — 'minor instances that could be trimmed', short of the every-token-earns-its-place anchor 5. | 4 / 5 |
Actionability | Instructions are generic template directives ("Clarify goals, constraints, and required inputs", "Apply relevant best practices and validate outcomes") with no formulas, code, or concrete steps for any metric; the single concrete pointer, `resources/implementation-playbook.md`, targets a file that does not exist in the bundle. This matches anchor 2 (high-level hints, missing specific steps) rather than 1 because the file pointer is at least a concrete navigation hint. | 2 / 5 |
Workflow Clarity | "Clarify goals → apply best practices → provide actionable steps" is a rough generic pseudo-sequence with many gaps — no ordered workflow for computing or validating any risk metric, and validation appears only as the abstract phrase "validate outcomes" with no checkpoint. This is a multi-method domain skill, so the simple-skill exception does not apply; it fits anchor 2, not 3, because the listed steps are template boilerplate rather than actual task steps. | 2 / 5 |
Progressive Disclosure | The body's structure depends entirely on one clearly-signaled reference ("If detailed examples are required, open `resources/implementation-playbook.md`") that is a dangling pointer — no resources/ directory (or references/, scripts/, assets/) exists in the bundle, so navigation dead-ends and the skill collapses to a 36-line stub. Fits anchor 2 (minimal effective structure); not 3, because there the references at least resolve. | 2 / 5 |
Total | 10 / 20 Passed |