Content
61%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 is a well-structured, actionable CLI usage guide with concrete commands and a sensible flow. Its main weaknesses are a marketing-flavored intro that adds tokens, the absence of validation/verification steps around destructive resource operations, and no worked end-to-end example.
Suggestions
Cut the marketing intro paragraph about what PublisherKit is for marketing teams; keep only what Claude needs to drive the CLI.
Add a validation/verification checkpoint for destructive actions (e.g., confirm the resource ID via a Get/list before Delete, and verify the returned status), which would also lift workflow_clarity above the destructive-cap of 3.
Include one concrete end-to-end example resolving the placeholders for a real action (e.g., listing and running a "Create Template" action with a sample --input JSON payload) to move actionability toward fully copy-paste-ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient command examples, but the opening marketing paragraph ("helps businesses manage and distribute their content across various channels... streamline their publishing workflows and reach wider audiences") and phrases like "so you can focus on the integration logic rather than auth plumbing" are unnecessary explanation that could be trimmed, matching the 'mostly efficient but includes some unnecessary explanation' anchor rather than the lean anchor 4. | 3 / 5 |
Actionability | It provides copy-paste-ready commands throughout ("npm install -g @membranehq/cli", "membrane search publisherkit --elementType=connector --json", "membrane action run --connectionId=CONNECTION_ID ACTION_ID --json", "membrane request CONNECTION_ID /path/to/endpoint") plus a flags table, but uses unresolved placeholders (CONNECTION_ID, ACTION_ID, QUERY) with no end-to-end worked example for a specific resource, fitting the 'mostly executable guidance with minor gaps' anchor. | 4 / 5 |
Workflow Clarity | A logical sequence is present (install → login → connect → search actions → run → proxy), but the skill exposes destructive operations (Create/Update/Delete Template, Campaign, Post) with no validation or verification checkpoints, and the rubric's destructive-operations cap at 3 applies and overrides any higher placement. | 3 / 5 |
Progressive Disclosure | No bundle files exist and the skill is a single well-organized SKILL.md with clear section headers and no nested references, giving good structure and easy navigation; at ~127 lines it exceeds the under-50-line simple-skill exception for a 5, so it lands at the 'good structure, most content appropriately placed' anchor. | 4 / 5 |
Total | 14 / 20 Passed |