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.
A solid CLI-integration skill: commands are executable, the connection state machine is handled with real checkpoints, and structure is sensible. Main costs are token slack (the bare entity list, the duplicated 'Popular actions' command) and a broken 'Step 2' cross-reference, plus inconsistent CLI invocation.
Suggestions
Cut or compress the 31-item bare entity list — either fold the entities into the description of what the skill covers, or move them to a references file with one line each on what to do with them.
Delete the 'Popular actions' section or populate it: as written it only repeats the 'Searching for actions' command and adds no actions.
Fix the dangling 'skip to **Step 2**' references by labeling the actual steps (or removing the numbering), and pick one CLI invocation form ('membrane' vs 'npx @membranehq/cli@latest') and use it consistently.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient — the Actindo intro is justified since it is niche knowledge — but it carries clear slack: a 31-item bare bullet list of entity names with no guidance attached, a 'Popular actions' section that only repeats the action-list command already shown, and a rambling Agent Types note ('etc. Those will be used to adjust tooling to be used best with your harness'). This fits 'Mostly efficient but includes some unnecessary explanation or could be tightened'. | 3 / 5 |
Actionability | Concrete, copy-paste-ready commands throughout (install, login, connection ensure, polling, action list/run with --input, proxy flags table), matching 'Mostly executable guidance; concrete code or commands with minor gaps'. Not 5 because of gaps: CLI invocation is inconsistent (bare 'membrane' vs 'npx @membranehq/cli@latest'), 'membrane login --tenant' leaves the tenant value unspecified, and the 'Popular actions' heading lists no actual actions. | 4 / 5 |
Workflow Clarity | The connect-wait-search-run-proxy sequence is clear and includes a genuine feedback loop (poll until state leaves BUILDING; branch on READY / CLIENT_ACTION_REQUIRED / CONFIGURATION_ERROR with an explicit 'poll again after the user acts' instruction), matching anchor 4. Not 5 because the body says 'skip to **Step 2**' twice but no Step 2 (or Step 1) heading exists — a dangling reference that breaks the numbered sequence. | 4 / 5 |
Progressive Disclosure | No bundle files exist, and the single-file body is organized into clear sections (Install, Authentication, Connecting, Searching, Running, Proxy, Best practices) fitting 'Good structure; most content is appropriately placed; minor organization gaps'. Not 5 because the 31-line bare entity list is inline content that earns little per token and the dangling 'Step 2' reference hurts navigation. | 4 / 5 |
Total | 15 / 20 Passed |