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.
The body is an efficient, actionable guide with a well-sequenced connection workflow and validation feedback loops. Main improvements are normalizing the CLI invocation, fixing step numbering, and moving the verbose clientAction-state detail into a reference file.
Suggestions
Normalize CLI invocation: pick either 'membrane' or 'npx @membranehq/cli@latest' consistently across all examples to avoid confusion.
Fix the step numbering so 'Step 1' and 'Step 2' are explicitly labeled rather than jumping to '1b' and referencing an unlabeled 'Step 2'.
Consider moving the detailed clientAction state handling and proxy flag table into a reference file, leaving SKILL.md as a tighter overview with a one-level-deep pointer.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly lean with concrete commands and flags, with only minor over-explanation such as 'so you can focus on the integration logic rather than auth plumbing' that could be trimmed. | 4 / 5 |
Actionability | It provides concrete, copy-paste-ready commands (install, login, connection ensure, action list/run, proxy) with a detailed flag table; minor gaps come from inconsistent 'membrane' vs 'npx @membranehq/cli' invocation and unresolved placeholders like CONNECTION_ID. | 4 / 5 |
Workflow Clarity | A clear install→authenticate→connect→poll→search/run sequence with validation checkpoints (poll until READY, inspect clientAction/error) and a feedback loop (re-poll after user action); step numbering is slightly confusing ('1b' and a referenced 'Step 2' without a labeled Step 1). | 4 / 5 |
Progressive Disclosure | Content is organized into clear, well-labeled sections with no nested references; with no bundle files present everything is inline, and at ~140 lines it sits just outside the simple-skill exception that would allow a 5. | 4 / 5 |
Total | 16 / 20 Passed |