Content
67%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.
The body delivers a concrete, executable workflow for connecting to Advanced Web Ranking via Membrane, with genuine state-polling feedback loops. Its main flaws are a dangling "Step 2" reference, an empty "Popular actions" section, and padded intro/boilerplate material.
Suggestions
Define or remove the "Step 2" reference: the text says "skip to Step 2" twice but no Step 2 exists; either number the sections consistently (1. Connect, 2. Discover actions, 3. Run) or drop the numbering.
Populate "Popular actions" with 3-5 actual Advanced Web Ranking actions (e.g. fetching a ranking report, listing keywords for a campaign) or remove the empty section header.
Trim the introductory paragraph and the "Use action names and parameters as needed" filler, and move the generic Membrane CLI install/login details to a shared reference file since they are common to all Membrane-based skills.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Commands are mostly presented leanly, but there is unnecessary padding: the intro paragraph explaining what Advanced Web Ranking is and who uses it, the content-free overview bullets ending in "Use action names and parameters as needed", and generic Membrane auth boilerplate that could be tightened. | 3 / 5 |
Actionability | Concrete, copy-paste-ready commands cover install, login (including headless completion), connection setup, action discovery, action runs with JSON input, and proxying with a flag table. Not a 5 because the "Popular actions" section is an empty header that never lists any actions, and one command inconsistently uses `npx @membranehq/cli` where others use `membrane`. | 4 / 5 |
Workflow Clarity | The sequence (install → login → ensure connection → poll state → search actions → run) is clear, with a real feedback loop for CLIENT_ACTION_REQUIRED, CONFIGURATION_ERROR, and polling until READY. Not a 5 because the body twice says "skip to Step 2" but no Step 2 section exists, and the "1b" numbering implies steps 1/2 that are not defined, leaving a dangling seam in the sequence. | 4 / 5 |
Progressive Disclosure | A single-file skill with no bundle files, well-organized section headers, and no nested references; the ~140-line body is a reasonable size for SKILL.md. Not a 5 because generic Membrane CLI/auth material (install, login, agent types) is duplicated inline and would sit better in a shared reference, and the empty "Popular actions" section adds navigation noise. | 4 / 5 |
Total | 15 / 20 Passed |