Content
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, highly actionable skill with excellent workflow clarity and specific API/tool guidance throughout. Its main weakness is length — at ~300+ lines with all CMS variants and diagnostic criteria inline, it could benefit from splitting reference material into separate files. The closing 'principles' section is somewhat redundant with guidance already woven into the workflow steps.
Suggestions
Extract the CMS-specific connection and push instructions (Steps 0 and 7) into a shared CMS_REFERENCE.md file, since this pattern appears in multiple skills (the skill itself references `prompt-gap-to-publish`).
Remove or significantly trim the 'What makes a page AI-cite-worthy' section — its principles are already operationalized in Steps 4 and 5, making the summary redundant.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is generally well-written but includes some unnecessary explanation, particularly the 'What makes a page AI-cite-worthy' section at the end which restates principles already embedded in the workflow steps. The CMS discovery table and per-CMS instructions in Step 7 are useful reference material but add significant length. Some prose could be tightened (e.g., the opening paragraph explains the logic Claude could infer). | 2 / 3 |
Actionability | The skill provides specific API calls (e.g., `get_ai_visibility_pages` with exact parameters like `sortBy: "citationCount"` and `mentionsBrandId`), concrete output formats (tables with specific columns), exact CMS tool names and parameters (e.g., `state: DRAFT` for HubSpot), and a clear section-level patch format with BEFORE/AFTER/WHY structure. The guidance is highly executable. | 3 / 3 |
Workflow Clarity | The 8-step workflow (Steps 0-7) is clearly sequenced with explicit checkpoints: Step 0 gates CMS discovery before analysis begins, Step 3 pauses for user selection, Step 4 requires user agreement on diagnosis before rewriting, Step 6 offers simulation validation before publishing, and Step 7 always saves as draft (never auto-publishes). The feedback loops and safety gates are well-placed throughout. | 3 / 3 |
Progressive Disclosure | The skill is a single monolithic file with no bundle files to offload detail into. The CMS-specific instructions in Steps 0 and 7, the diagnosis dimensions in Step 4, and the 'What makes a page AI-cite-worthy' section could all be split into referenced files. There's one reference to the `prompt-gap-to-publish` skill for CMS setup guidance, but otherwise everything is inline. | 2 / 3 |
Total | 10 / 12 Passed |