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 is a solid, actionable runbook: every stage of the Membrane workflow has copy-paste commands and the connection lifecycle includes genuine validation and error-recovery guidance. Its weaknesses are moderate padding (filler sentence, empty table column, unexplained resource tree) and a dangling 'Step 2' reference that breaks the numbered flow.
Suggestions
Delete the filler sentence "Use action names and parameters as needed." and either fill in the empty Description column of the Popular actions table or drop that column.
Fix the dangling navigation: replace "skip to Step 2" and the lone "1b." heading with explicitly numbered steps (Step 1: connect, Step 2: search and run actions) or reword them to point to the actual section headings.
Trim or annotate the 15-item resource-type tree — as a bare list it adds tokens without telling the reader when to care about Contacts, Pipelines, Teams, etc.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient — the CLI commands and connection-state handling earn their tokens — but there is clear padding: "Use action names and parameters as needed." is pure filler, the resource tree lists 15+ items with no accompanying guidance, the Popular actions table has an entirely empty Description column, and "Agent Types : claude, openclaw, codex, warp, windsurf, etc." is loose prose. Fits anchor 3 ('Mostly efficient but includes some unnecessary explanation or could be tightened'); not 2 because the core instructional content is genuinely useful, not padded. | 3 / 5 |
Actionability | Concrete copy-paste commands cover the whole workflow: "npm install -g @membranehq/cli@latest", "membrane login --tenant --clientName=<agentType>", "membrane connection ensure \"https://www.apivoid.com/\" --json", "membrane action list --connectionId=CONNECTION_ID --intent \"QUERY\"", "membrane action run <actionId> --input '{\"key\": \"value\"}'", plus a flags table for proxy requests. Not 5: no end-to-end example with real values (e.g. actually running check-url-reputation against a URL), and the Popular actions table's empty Description column leaves the reader not knowing what each action does — minor gaps consistent with anchor 4. | 4 / 5 |
Workflow Clarity | A clear sequence (install → login → connection ensure → poll until READY → handle client actions → search actions → run) with real checkpoints: the --wait long-poll, "poll again with membrane connection get <id> --json to check if the state moved to READY", and explicit CONFIGURATION_ERROR/SETUP_FAILED handling form a validate-and-recover loop. Not 5 because "If the returned connection has state: READY, skip to Step 2" references a Step 2 that is never labeled, and the "#### 1b." heading exists with no Step 1 or Step 2 — dangling navigation that keeps it at anchor 4's 'minor gaps'. | 4 / 5 |
Progressive Disclosure | No bundle files exist (no references/, scripts/, assets/), so this is a single-file skill of ~160 lines with well-organized, clearly headed sections (Install, Authentication, Connecting, Searching, Running, Proxy, Best practices). That matches anchor 4's 'Good structure; most content is appropriately placed... minor organization gaps' — the CLI flag reference, connection-state guide, and action table could arguably live in separate reference files, and the content-free resource tree adds navigational noise, which keeps it below 5. | 4 / 5 |
Total | 15 / 20 Passed |