Content
75%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.
A well-structured integration guide with executable commands, explicit connection-state feedback loops, and a sensible single-file layout sized to the skill. The main weaknesses are a dangling 'Step 2' reference, an empty 'Popular actions' section that promises but does not deliver common cases, and inconsistent CLI invocation (membrane vs npx @membranehq/cli).
Suggestions
Fix the dangling reference: 'skip to Step 2' points to a step that is never defined — either number the sections or reword to 'skip to Searching for actions'.
Populate 'Popular actions' with 3-5 concrete worked examples (e.g. searching candidates, creating a placement) instead of repeating the discovery command, and pick one CLI invocation (membrane vs npx @membranehq/cli) and use it consistently.
Trim non-actionable filler such as 'Use action names and parameters as needed.' and the URL-normalization mechanics.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is command-driven and mostly lean, and the explanations of Membrane connection states are genuinely unknown operational detail. A few tokens do not earn their place — the vacuous 'Use action names and parameters as needed.', the 'Popular actions' section that only repeats a command, and mechanics like 'The URL is normalized to a domain and matched against known apps' — which is slightly more than 'minor instances', keeping it below the lean level-5 anchor. | 4 / 5 |
Actionability | Copy-paste-ready commands cover the whole flow ('npm install -g @membranehq/cli@latest', 'membrane connection ensure "https://ceipal.com" --json', 'membrane action run <actionId> --input ... --json') plus a proxy flag table. It falls short of the level-5 anchor because 'specific examples cover the common cases' is not met — the 'Popular actions' section lists no actual actions, there is no worked end-to-end example, and invocation alternates between 'membrane' and 'npx @membranehq/cli@latest'. | 4 / 5 |
Workflow Clarity | The sequence install -> login (with headless fallback) -> connection ensure -> poll states -> search actions -> run is clear, with strong error-state handling (CLIENT_ACTION_REQUIRED sub-actions, CONFIGURATION_ERROR, and a re-poll feedback loop after the user authenticates). It misses level 5 because of a real coherence gap: 'skip to Step 2' references a step label that does not exist anywhere in the document. | 4 / 5 |
Progressive Disclosure | Scored against the actual bundle structure — no references/, scripts/, or assets/ directories exist and the body cites no separate files, so nothing is nested or buried. The single ~180-line file with clear section headers is appropriately placed for this skill's size; minor gaps (the mislabeled 'Popular actions' section) keep it below the level-5 'well-signaled references' anchor, which is inapplicable here but reflects ideal organization. | 4 / 5 |
Total | 16 / 20 Passed |