Content
67%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 delivers strongly executable CLI guidance with a well-handled connection-state feedback loop, and its structure is easy to navigate as a single-file skill. Its weaknesses are moderate verbosity (marketing filler and a vague overview line) and a broken step reference ('Step 2' is never defined) plus inconsistent CLI invocation styles.
Suggestions
Delete the marketing opener and filler sentences ('helps professionals and teams manage their time more efficiently', 'so you can focus on the integration logic rather than auth plumbing') and the empty line 'Use action names and parameters as needed.'
Fix the broken workflow numbering: 'skip to Step 2' references a heading that doesn't exist — rename '### Searching for actions' to '### Step 2: Searching for actions' or remove the step references.
Standardize CLI invocation (always 'membrane ...' or always 'npx @membranehq/cli ...') so commands are consistently copy-paste runnable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly command-driven and efficient, but includes padding Claude doesn't need: the marketing opener ('helps professionals and teams manage their time more efficiently'), 'so you can focus on the integration logic rather than auth plumbing', and the contentless line 'Use action names and parameters as needed.', fitting the 'mostly efficient but includes some unnecessary explanation' anchor. | 3 / 5 |
Actionability | Nearly all guidance is copy-paste executable (install, login, connection ensure/get, action list/run, proxy flags table), but minor gaps remain: the undefined 'Step 2' reference, inconsistent 'membrane' vs 'npx @membranehq/cli' invocation styles, and one vague overview line, keeping it below the fully-executable anchor. | 4 / 5 |
Workflow Clarity | The sequence (install → authenticate → connect → poll for readiness → search actions → run/proxy) is clear with a genuine feedback loop for BUILDING and CLIENT_ACTION_REQUIRED states, but the text tells the reader to 'skip to Step 2' when no Step 2 heading exists, and there is no validation before running actions that could mutate data. | 4 / 5 |
Progressive Disclosure | The single-file skill is well-sectioned with clear headers, an external docs link, and appropriately inline tables; since no bundle files exist there is nothing to offload, but minor organization gaps (dangling step numbering, a redundant 'Popular actions' intro repeating the search command) keep it at the good-structure anchor rather than the clear-overview anchor. | 4 / 5 |
Total | 15 / 20 Passed |