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.
The body is highly actionable with copy-paste-ready CLI commands and a clear connection-flow walkthrough, and is reasonably well-structured. Its weaknesses are modest over-explanation in spots and the absence of validation/verification steps around destructive and batch ad operations.
Suggestions
Add an explicit validation/verification checkpoint after create/update/delete actions (e.g. re-list or get the campaign to confirm the change took effect) to raise workflow clarity past the destructive/batch cap.
Tighten the overview paragraph and the clientAction state explanations — Claude can infer OAuth/auth-reconnect semantics; keep only the Membrane-specific behavior.
Consider moving the full popular-actions table and proxy flag reference into separate reference files (e.g. ACTIONS.md) to deepen progressive disclosure while keeping SKILL.md a lean overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | It is mostly efficient with concrete commands and tables, but the overview paragraph and the 'clientAction' state walkthrough over-explain mechanics Claude could infer, and some sections (proxy flag tables, best-practice rationale) could be trimmed. | 3 / 5 |
Actionability | It provides fully executable, copy-paste-ready commands throughout (install, login, connection ensure/get, action list/run, request with flag tables) covering the common cases with concrete examples and the --json output convention. | 5 / 5 |
Workflow Clarity | A rough connection lifecycle sequence is present with state checks (READY/BUILDING/CLIENT_ACTION_REQUIRED), but operations that are destructive or batch in nature (delete/update campaigns) lack explicit validation or verification checkpoints, capping workflow clarity at 3 per the destructive/batch rule. | 3 / 5 |
Progressive Disclosure | Content is well-organized into clear sections with a popular-actions table and CLI subcommands grouped logically; there are no bundle files so it stays self-contained and shallow, though a couple of inline reference blocks (the large actions table, proxy flags) could arguably live in separate reference files. | 4 / 5 |
Total | 15 / 20 Passed |