Content
82%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 content is highly actionable, with copy-paste-ready commands and a clear install-to-execution workflow that includes real error-recovery loops in the connection phase. The main gaps are minor conciseness padding (the bare Overview bullets) and missing result-validation guidance for the run/proxy steps.
Suggestions
Replace the 15 bare-bullet Overview and the filler "Use action names and parameters as needed." with a one-line purpose statement, or fold the bullet list into a short note pointing to action discovery via `membrane action list`.
Add a short validate-after-run note for `membrane action run` and `membrane request`, e.g. checking the response `error` field or non-2xx status before proceeding, to close the validation gap in the batch/proxy steps.
Trim authoring-voice sentences like "so you can focus on the integration logic rather than auth plumbing" that restate what the command already demonstrates.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is largely executable commands with little padding, but it includes minor over-explanation ("Membrane handles authentication... so you can focus on the integration logic rather than auth plumbing") and a 15-item bare-bullet Overview plus the filler line "Use action names and parameters as needed," which could be trimmed — efficient with minor instances of over-explanation rather than the lean every-token-earns-its-place level of a 5. | 4 / 5 |
Actionability | It provides copy-paste-ready `membrane` commands with real flags (--json, --wait, --input, --method, --header, --query) covering the common cases — install, login, connect, search actions, run actions with JSON input, and proxy requests — plus a popular-actions table mapping names to keys, matching 'fully executable; specific examples cover the common cases'. | 5 / 5 |
Workflow Clarity | The install→login→connect→search→run sequence is clear, and the connection phase has genuine validation feedback loops (poll --wait, branch on READY/CLIENT_ACTION_REQUIRED/CONFIGURATION_ERROR, re-poll after user action), but the later 'Running actions' and 'Proxy requests' steps lack explicit error/result validation, leaving minor validation gaps that keep it below a 5. | 4 / 5 |
Progressive Disclosure | The single-file SKILL.md is well-organized with clear section headers and a navigable structure (install, auth, connecting, searching, popular actions, running, proxy, best practices); with no bundle files and content mostly appropriately placed for a skill of this size it has good structure, though the inlined popular-actions and proxy-flag tables and the bare Overview bullets are minor organization gaps rather than the cleanly one-level-deep referenced structure of a 5. | 4 / 5 |
Total | 17 / 20 Passed |