Content
81%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 strong, executable skill body with copy-paste-ready commands, clear argument documentation, and a working bundle structure. Its main weakness is token efficiency: the Interpretation sections re-explain standard technical-analysis and statistics concepts Claude already knows, and global sections (Dependencies, Timezone) are placed confusingly at the end under the Correlation section.
Suggestions
Trim the Interpretation sections to skill-specific thresholds only (e.g., 'crossover.days_ago of 0-5 = fresh signal', Sharpe/volatility cutoffs the script reports) and drop textbook facts Claude already knows like 'RSI > 70 = overbought', 'Golden cross (SMA20 > SMA50) = bullish', and the correlation-near-1.0/-1.0/0 explanations.
Move the global '## Dependencies' and '## Timezone' sections to a top-level position (before the per-tool sections) so their scope clearly applies to both tools, and keep each tool's section self-contained.
Remove the duplicated stock-scoping between the intro line and frontmatter description (e.g., 'Supports multi-symbol analysis and earnings data' can be stated once).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient, but the Interpretation sections spend ~10 lines on textbook concepts Claude already knows: 'RSI > 70 = overbought', 'Golden cross (SMA20 > SMA50) = bullish', and 'Correlation near 1.0 = highly correlated (move together)'. Only skill-specific thresholds (crossover days_ago freshness, Sharpe cutoffs, timezone requirements) earn their tokens. | 3 / 5 |
Actionability | Commands are fully executable and copy-paste ready (e.g., 'uv run python scripts/technicals.py AAPL,MSFT,GOOGL'), examples cover the common cases (single, multi-symbol, earnings, indicator subsets, correlation), and the uv/python fallback note handles environment variance. | 5 / 5 |
Workflow Clarity | Each task is a single unambiguous command with documented arguments and outputs, and the scripts are read-only analysis so no destructive/batch validation cap applies. The simple-skill exception is satisfied: the single action per invocation is completely clear. | 5 / 5 |
Progressive Disclosure | Good structure with both referenced scripts (scripts/technicals.py, scripts/correlation.py) actually present in the bundle, and content appropriately inline for a skill this size. Minor gaps: the global '## Dependencies' and '## Timezone' sections sit at the bottom under the Correlation section despite applying to both tools, and the Timezone section's placement makes its scope ambiguous. | 4 / 5 |
Total | 17 / 20 Passed |