Content
57%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 skill body is actionable and well-structured with executable code and clear signal thresholds, but it is a lengthy single-file document with no progressive disclosure to deeper references and only an implicit analysis workflow.
Suggestions
Add an explicit numbered analysis workflow (e.g., 1. Pull supply/mint-burn data, 2. Compute per-dimension scores, 3. Aggregate composite, 4. Render output) with a verification/sanity-check step before emitting the report.
Move the chain-level distribution tables and data-source reference into a separate references file (e.g., references/data-sources.md) and link to it from SKILL.md to enable progressive disclosure.
Trim the opening overview paragraph and interpretive prose to the essential signal definitions to improve token efficiency.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient code and threshold tables, but the opening 'dry powder' paragraph and several interpretive prose blocks (e.g., 'USDC is more regulated and institutional-oriented') could be tightened, fitting anchor 3 rather than 4. | 3 / 5 |
Actionability | Provides concrete, executable Python functions (mint_burn_signal, exchange_reserve_signal, velocity_signal) with explicit thresholds plus a composite-score structure and output template; the gap is that data-fetching is only described via a sources table rather than shown in code, keeping it just below anchor 5. | 4 / 5 |
Workflow Clarity | The seven numbered core concepts plus the output-format template imply a sequence, but there is no explicit step-by-step workflow or validation checkpoints; this is an analytical (non-destructive) skill so the destructive-cap does not apply, landing it at anchor 3. | 3 / 5 |
Progressive Disclosure | No bundle files exist and all content lives inline in a >50-line monolithic SKILL.md; sections are well-organized but content that could be split into references (per-chain tables, data-source details) is inlined, matching anchor 3 rather than 4. | 3 / 5 |
Total | 13 / 20 Passed |