Content
53%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.
A well-organized, domain-appropriate instruction skill with a useful R-multiple framework and a concrete expectancy formula, but it is held back by generic boilerplate sections, missing computational specifics for position sizing/VaR/hedging, and no verification checkpoints in its workflow. The structure is clean and appropriately self-contained.
Suggestions
Remove or replace the generic 'Instructions' and 'Use/Do not use this skill when' boilerplate with domain-specific content — e.g., specify the actual account-risk percentage range and the position-sizing formula (risk amount / stop distance).
Add executable detail to the vague directives: a concrete VaR calculation method, a monte carlo parameter setup, and an example hedging structure instead of 'Use monte carlo simulations for stress testing'.
Insert a validation checkpoint into the Approach workflow — e.g., 'Verify total portfolio heat (sum of open R) stays below the documented limit before recommending new positions' — so the sequence has an explicit verification step.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The Focus Areas, Approach, and Output sections are lean bullet lists that assume Claude's competence (no explaining what VaR or Kelly criterion is), but roughly a fifth of the body is generic boilerplate — 'Clarify goals, constraints, and required inputs', 'Apply relevant best practices and validate outcomes', and the circular 'Use this skill when working on risk manager tasks' — that could be trimmed. Mostly efficient with unnecessary filler, matching anchor 3 rather than anchor 4's 'minor instances'. | 3 / 5 |
Actionability | Some concrete guidance exists — the expectancy formula '(Win% × Avg Win) - (Loss% × Avg Loss)' and the '1R = max loss' definition — but key details are missing: no position-sizing formula, no method for the VaR or monte carlo calculations, and 'Size positions based on account risk percentage' gives no percentage or computation. As an instruction-only skill this is actionable in outline but incomplete, fitting anchor 3; not 2 because the formula and R-framework are genuinely executable. | 3 / 5 |
Workflow Clarity | The Approach section gives a clear 7-step numbered sequence, but validation checkpoints are absent or only implicit — 'Document risk limits and stick to them' is a directive, not a checkpoint, and no step verifies computed risk totals or outputs. Not 4 because most checkpoints are missing rather than minor gaps; not 2 because the sequence itself is well-defined and coherent. | 3 / 5 |
Progressive Disclosure | The body is self-contained with well-organized sections (Use when, Instructions, Focus Areas, Approach, Output) and no buried or nested references — appropriate for a guidance skill. It sits just over the ~50-line simple-skill threshold and topics like VaR or monte carlo stress testing could arguably be split into reference files, so anchor 4's 'good structure; minor organization gaps' fits better than anchor 5. | 4 / 5 |
Total | 13 / 20 Passed |