Content
85%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.
An unusually strong operational skill: concrete tool calls, cost guardrails, validation checkpoints, and an exactly-specified output format including copy-paste-ready SVG and JSON examples. Its one structural weakness is that it is a monolith — the detailed rank-grid drawing specification could be split into a one-level-deep reference file, and a few minor repetitions could be trimmed.
Suggestions
Move the "Drawing the grid" section (color scale, SVG template, stat blocks, widget JSON format) into a reference file such as references/rank-grid.md and point to it from the "Maps coverage" output step, keeping SKILL.md as the overview.
State the prioritization rule ("category and claim problems outrank posting cadence") once in the workflow and reference it from the output format instead of repeating it verbatim.
Tighten a few long sentences (e.g., the search_local_businesses bullet mixing prospecting guidance with chain-coverage guidance) to recover tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and operational — tool semantics, cost rules, and output specs that Claude would not already know — with almost no padding or explanation of known concepts (e.g., "Local rank trackers (SE Ranking, Local Falcon, BrightLocal) all draw the same picture" is genuinely novel context). Minor trimming is possible (the rule "Category and claim problems outrank posting cadence every time" is stated twice, and some sentences run long), which keeps it below the lean-every-token-earns-its-place anchor at 5 but clearly above the midpoint. | 4 / 5 |
Actionability | Guidance is fully executable: named MCP tools with concrete parameters ("isClaimed: false", "search_local_serp_results ... near the business coordinate"), a copy-paste-ready SVG template for the 3x3 grid with exact hex colors and a JSON widget example, and a fully specified report section order. Specific examples cover the common cases, matching the top anchor. | 5 / 5 |
Workflow Clarity | The 8-step workflow is clearly sequenced with explicit validation checkpoints: "confirm the coordinate matches the storefront before spending grid credits", "state the cost first" before multi-keyword grids, sanity-checking website links in step 4, and a polling feedback loop for queued reviews ("call again with it after 30-60 seconds"). The paid-batch grid operation is properly guarded, satisfying the validation cap. | 5 / 5 |
Progressive Disclosure | The body is well-sectioned with clear headers, but everything lives in one 137-line SKILL.md with no reference files at all. The ~55-line "Drawing the grid" spec (SVG template, color table, stat blocks, widget JSON format) is detailed material that belongs in a one-level-deep reference file, matching the anchor 'content that should be separate is inline'. Not a 2 because structure is strong and nothing is buried, and not a 4 because a substantial, separable spec is inlined in the overview file. | 3 / 5 |
Total | 17 / 20 Passed |