Content
86%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 tight, highly actionable body: executable CLI invocations for all four commands, a useful Rules section capturing real response quirks, and clean progressive disclosure to a single reference file. The only refinements are minor redundancy in the Overview and an implied (not explicit) search-to-detail chaining step.
Suggestions
Drop or merge the Overview's four-capability bullets with the 'When to Use This Skill' table to remove the capability list stated twice in close succession.
Add a one-line workflow note that 'search' returns a contractAddress to pass into 'meta', 'dynamic', and 'kline', making the cross-command flow explicit instead of implied.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body uses compact tables and a concrete Rules section with no concept-padding Claude already knows, but the Overview paragraph restates the four capabilities that the 'When to Use This Skill' table immediately repeats — a minor instance of over-explanation that could be trimmed. | 4 / 5 |
Actionability | Every command ships a copy-paste-ready invocation ('node <skill-dir>/scripts/cli.mjs search '{"keyword":"<keyword>","chainIds":"1,56,8453,CT_501"}'') with required args and real parameter values, covering all four common cases. | 5 / 5 |
Workflow Clarity | Each subcommand is an unambiguous single read-only action and intent is mapped to command in a table, but the cross-command pattern (use 'search' to obtain a contractAddress, then feed it to 'meta'/'dynamic'/'kline') is only implied rather than sequenced — a minor gap. | 4 / 5 |
Progressive Disclosure | SKILL.md is a lean overview that defers per-subcommand invocations, parameter tables, return-field tables, and response samples to a clearly signaled, one-level-deep references/cli.md (a real bundle file), with scripts/cli.mjs as the executable — matching clear overview with well-signaled one-level references. | 5 / 5 |
Total | 18 / 20 Passed |