Content
82%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 dense, high-signal knowledge skill: concrete symbol anchors, exact message contracts, a sequenced change checklist, and repo-specific verification commands with almost no wasted tokens. The main gaps are shape-only message examples and the absence of an explicit error-recovery loop and externalized reference material for the message contracts.
Suggestions
Add a short fix-and-retry loop after verification steps, e.g. 'If agent:check or lint fails, fix the reported errors in the touched files and re-run before committing.'
Move the detailed per-message contract sections (kline, marks, price scale, symbol sync, perps lines) into a references/ file with one-level-deep links, keeping SKILL.md as the mental-model and checklist overview.
Flesh out at least the most common message snippets (e.g. tradingview_getKLineData) with representative values or types so they are copy-paste adaptable rather than field-name-only.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and dense — search anchors, message field lists, and a change checklist consisting entirely of project-specific facts with zero explanation of concepts Claude already knows; every token earns its place, matching anchor 5 rather than the minor-trim case of anchor 4. | 5 / 5 |
Actionability | Concrete, executable guidance dominates: grep-able symbol anchors ("useTradingViewUrl", "setupSymbolChangeListener"), specific commands ("yarn agent:check --profile commit", "localhost:5173"), and exact message fields; the message-contract snippets are shape-only (field names without values or types), a minor gap keeping it below the fully copy-paste anchor 5 and above the pseudocode anchor 3. | 4 / 5 |
Workflow Clarity | The "Change Checklist" gives a clear 9-step sequence with explicit verification ("run the relevant TypeScript/lint command", "npm run build", "yarn lint") and the Verification section adds repo-specific checks; it lacks an explicit fix-and-retry feedback loop for failed checks, keeping it below anchor 5 and above anchor 3 where validation is absent. | 4 / 5 |
Progressive Disclosure | The body is well-organized with clear section headers and no nested references, and no bundle files exist to mis-navigate; the detailed per-message contract sections (kline, marks, price scale, symbol sync, perps lines) could arguably be split into a one-level-deep reference file, a minor organization gap matching anchor 4 rather than the well-split anchor 5. | 4 / 5 |
Total | 17 / 20 Passed |