Content
71%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 compact, well-structured skill body with a concrete routing table, a defined output schema, explicit cost guardrails, and a prerequisite validation checkpoint — but it never shows how to actually execute a query (API endpoint/parameters or the underlying CLI call), which is the main actionability gap. Progressive disclosure and conciseness are strong; error-handling validation is the secondary weakness.
Suggestions
Show the concrete invocation for the core task — the SE Ranking API endpoint (with method and key parameters) or the exact `claude-seo`/CLI command — so the ai-visibility, serp, backlinks, and competitors routes are executable rather than just named.
Add a short error-handling checkpoint: what to do on API failure, unit exhaustion, or empty/low-sample responses before reporting results.
Trim the vendor-comparison trivia ("same logic used by Profound / Peec AI") unless it changes a routing decision.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean — routing table, output-field bullets, cost units, and delegation pointers with almost no padding and no explanations of concepts Claude already knows. Not a 5 because sentences like "The scorer is the same logic used by Profound / Peec AI but bundled into one MCP/API" are contextual trivia that could be trimmed. | 4 / 5 |
Actionability | Concrete elements exist — the `/seo seranking <command>` routing patterns, named output fields, the API-key check in `~/.claude/settings.json`, and the cost-logging command — but the actual execution mechanism for the queries (SE Ranking endpoint, parameters, or the CLI invocation for `ai-visibility`) is never shown, leaving key details missing. Not a 4 because the gap is the core action itself, not a minor one; not a 2 because the command surface and reporting requirements are specific. | 3 / 5 |
Workflow Clarity | A clear implicit sequence — install, verify `SERANKING_API_KEY` (with an explicit failure path: tell the user to run the installer), execute the routed command, report each SOV field "as a percentage with a confidence note based on sample size" — covers most checkpoints. Not a 5 because there is no handling for API errors or unexpected/empty responses, and no validation that the query output matches the expected fields. | 4 / 5 |
Progressive Disclosure | At ~44 lines this is a simple skill with no need for external reference files: it uses well-organized sections (Prerequisites, Routing, scoring fields, Cost guardrails, Cross-skill delegation), keeps the field list appropriately inline, and delegates deep-dives to sibling skills one hop away. No nested references or inlined bulk content. | 5 / 5 |
Total | 16 / 20 Passed |