Content
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a well-structured, information-rich reference with explicit thresholds and a usable report template, but it is monolithic, omits the executable data-acquisition step, and frames the analysis workflow only implicitly. Splitting reference material into bundle files and adding an explicit step-by-step procedure would lift the weaker dimensions.
Suggestions
Add a short executable example for fetching on-chain data (e.g., a Glassnode/Nansen API call) so the analysis pipeline is actionable end-to-end, not just thresholds.
Pull the detailed historical-reference tables and per-metric deep dives into separate reference files (e.g., references/metrics.md, references/history.md) and link them from SKILL.md to improve progressive disclosure.
Add an explicit numbered "Analysis procedure" section (gather snapshot → score per dimension → cross-check whale/capital flow → write report) with a verification checkpoint confirming data freshness before concluding.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense, mostly non-redundant reference data, but it runs ~250 lines and includes some removable hand-holding ("Simply put: on average, are the coins sold today being sold at a profit or a loss?", "Similar to a PE ratio") and speculative historical tables that could be trimmed. | 2 / 3 |
Actionability | It provides concrete thresholds, formulas, and a copy-paste report template, but the actual data-acquisition step is missing — it only names external APIs (Glassnode/Nansen) without any executable call, leaving a key detail incomplete. | 2 / 3 |
Workflow Clarity | The composite-scoring framework and output template imply a sequence (collect metrics → score → report), but the analysis is never laid out as explicit numbered steps with checkpoints; for a multi-step analytical workflow the sequence stays implicit. | 2 / 3 |
Progressive Disclosure | Sections are well-organized, but everything is in one monolithic SKILL.md with no bundle files (references/scripts/assets absent) and no pointers to separate references, so content that could be split (historical tables, per-metric deep-dives) is inline. | 2 / 3 |
Total | 8 / 12 Passed |