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 actionable and well-structured with concrete CLI commands and a clear connection-readiness workflow, but it over-explains in places and omits validation before potentially destructive action runs. As a single-file skill it is easy to navigate.
Suggestions
Trim the marketing-style intro paragraph and 'auth plumbing' phrasing to remove tokens that restate what Claude already knows.
Add a validation/verification step before running irreversible actions such as sending campaigns (e.g., confirm with the user or perform a dry-run/preview first).
Include one fully-worked end-to-end example with concrete values filled in, instead of relying solely on placeholders like CONNECTION_ID and QUERY.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient with concrete commands, but the introductory app-description paragraph and phrases like 'so you can focus on the integration logic rather than auth plumbing' over-explain what Claude can already infer. | 3 / 5 |
Actionability | Provides concrete, executable `membrane` CLI commands for install, auth, connection setup, action discovery, and proxying, plus a flag table; only placeholder gaps (CONNECTION_ID, QUERY, actionId) keep it from a 5. | 4 / 5 |
Workflow Clarity | Connection setup has polling/validation checkpoints (READY/BUILDING/CLIENT_ACTION_REQUIRED handling), but running actions—which can send irreversible newsletters—has no validation or verification before execution, capping the score per the destructive-operation rule. | 3 / 5 |
Progressive Disclosure | A single, well-organized SKILL.md with clear section headers and no nested or broken references, though all content is inline since no reference files exist. | 4 / 5 |
Total | 14 / 20 Passed |