Content
88%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, well-structured CLI skill body: efficient token use, a complete command table with runnable examples, and clear intent-to-command mapping appropriate to a simple single-purpose tool. The one material weakness is that the referenced scripts/yf file is missing from the bundle, undermining both executability and the body's only file reference.
Suggestions
Include the actual yf script in the bundle under scripts/ so the referenced {baseDir}/scripts/yf path resolves and the commands are genuinely executable.
Consolidate the duplicate --json note (stated in both the Commands and Output sections) into one place.
Make Error Handling concrete: specify the exit behavior or error-message format the script produces and an explicit retry delay for rate limits, rather than the vague "handles... gracefully".
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean: a setup command, a compact 11-row command table with one-line examples, a short "When to Use Which Command" decision list, and two brief closing notes. It assumes Claude's competence (no explanation of what Yahoo Finance or Rich tables are). Only trivial trimmable redundancy exists (the --json flag is mentioned in both Commands and Output), which the 5 anchor's "every token earns its place" still tolerates. | 5 / 5 |
Actionability | Every command has a concrete, copy-paste-ready example ("yf price YPF", "yf compare YPF,PAM,GGAL") and setup is a single executable chmod — but the entire skill hinges on "{baseDir}/scripts/yf", and no scripts/ directory exists in the bundle, so the core artifact the guidance points to cannot actually be executed. Concrete commands with a material gap fits the "mostly executable guidance; minor gaps" anchor rather than 5. | 4 / 5 |
Workflow Clarity | This is a simple, single-purpose skill (run one yf subcommand), and each action is unambiguous with a worked example; the "When to Use Which Command" section maps intents to commands. Operations are read-only queries, so no validation checkpoints are required, and the simple-skill exception applies. | 5 / 5 |
Progressive Disclosure | The body is well organized into Setup / Commands / When to Use / Error Handling / Output with no inline bloat that belongs in separate files, and at ~45 lines it needs no external references. However, scored against the actual bundle structure, its single referenced path ({baseDir}/scripts/yf) is absent from the bundle, leaving the file structure incomplete relative to what the body signals — a minor organization gap at anchor 4 rather than a clean 5. | 4 / 5 |
Total | 18 / 20 Passed |