Content
92%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, highly actionable skill body: executable commands throughout, a validated setup workflow with error recovery, and clean one-level-deep progressive disclosure into references/commands.md. The only flaw is redundancy — duplicated troubleshooting tables and repeated read-only/status statements that inflate token cost without adding information.
Suggestions
Collapse the Step 1 'Common setup issues' table and the end-of-file 'Error Reference' table into one troubleshooting section, since five of their rows are near-duplicates.
State the read-only guarantee once prominently (e.g. the bold callout in the intro) and trim its repeats in Key rule 10 and the headnote.
Merge the Step 5 Diagnostics guidance into Key rule 1's 'run status first' instruction to remove the third prescription of `opencli tradingview status`.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is table-driven and assumes Claude's competence with no concept padding, but the Step 1 "Common setup issues" table nearly duplicates the end-of-file "Error Reference" table, the read-only guarantee is stated three times, and `opencli tradingview status` is prescribed in Step 1, Key rule 1, and Step 5. These are minor trimmable redundancies — anchor 4, not the every-token-earns-its-place anchor 5. | 4 / 5 |
Actionability | Every example is copy-paste ready with real flags and values (e.g. `options-chain --ticker SNDK --expiry 2026-05-22 -f json`, the RSI<30 screener with a literal `--filter` JSON clause), and per-command output columns plus the `-f` format table cover the common cases fully. | 5 / 5 |
Workflow Clarity | A clear 5-step sequence with explicit validation (the `!` status probe branching on READY/SETUP_NEEDED) and error-recovery feedback loops ("If CDP is down, run `opencli tradingview launch`", plus the error→cause→fix table). The skill is read-only, so the destructive/batch validation cap does not apply. | 5 / 5 |
Progressive Disclosure | The body is an overview with the single deep-dive (references/commands.md, verified to exist and be one level deep) clearly signaled both inline in Key rule 6 and in a dedicated "Reference Files" section. Content is appropriately split between the SKILL.md quick tables and the reference's full flag documentation. | 5 / 5 |
Total | 19 / 20 Passed |