Content
68%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 content is well-structured and highly actionable with concrete Membrane CLI commands and a useful action table, and is reasonably concise. Its main weakness is workflow clarity: destructive and batch operations (delete/update by IDs) ship with no validation or verification checkpoints, capping that dimension.
Suggestions
Add a validation/verification checkpoint before destructive and batch actions (e.g., confirm IDs via a get/list query, run deletes with --dry-run or against a single ID first, and verify the result) so destructive workflows clear the workflow_clarity cap.
Trim the opening platform-marketing paragraph ('Crowd.dev is a community growth platform that helps companies...') to a one-line orientation, since Claude does not need the audience/use-case detail to execute the skill.
Consider moving the Popular actions table into a references/ file referenced one level deep from SKILL.md to tighten the overview and lift progressive_disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly command- and table-driven and efficient, but the opening paragraph explaining what Crowd.dev is and who uses it ('helps companies understand, engage, and grow their developer communities... used by developer relations teams...') is over-explanation that could be trimmed, fitting 'efficient; minor instances of over-explanation'. | 4 / 5 |
Actionability | Provides concrete, copy-paste-ready commands (npm install, membrane login, membrane search/connect/action list/action run/request) with flags and a Popular actions table of real action IDs, matching 'mostly executable guidance; concrete commands with minor gaps'; placeholders like CONNECTION_ID and the vague 'Use action names and parameters as needed' prevent a 5. | 4 / 5 |
Workflow Clarity | A clear sequence exists (install → login → connect → list actions → run/proxy), but the skill exposes destructive and batch operations ('Delete Organizations', 'Delete Members', 'Update Organization') with no validation, confirmation, or verify-result checkpoints; per the rubric cap, missing validation for destructive/batch operations caps this at 3. | 3 / 5 |
Progressive Disclosure | Content is organized into well-signaled sections (Install, Setup, Connecting, Searching, Popular actions, Running, Proxy, Best practices) with no nested references, matching 'good structure; most content appropriately placed; minor organization gaps'; the inline ~18-row action table is bulk content that could live in a separate reference file, holding it below 5. | 4 / 5 |
Total | 15 / 20 Passed |