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 a strong, execution-first integration guide: concrete CLI commands throughout, a complete connection-state decision tree, and a useful action catalog. Its main flaws are minor — a few marketing-fluff sentences, a 'Step 2' reference to a step that is never labeled, and bulky inline tables that could be split into reference files for a leaner SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dominated by executable commands and compact tables, but contains a few trimmable passages: the marketing-teams overview ('Marketing teams and sales organizations use it to optimize their advertising spend'), 'This is the fastest way to get a connection', and 'so you can focus on the integration logic rather than auth plumbing'. Anchor 4 ('efficient; minor instances of over-explanation') fits — anchor 5 requires every token to earn its place. | 4 / 5 |
Actionability | Everything is copy-paste ready: install command, login (including headless completion with `membrane login complete <code>`), `connection ensure` with a real URL, `action list`/`action run` with `--input` JSON syntax, a proxy command with a full flag table, and an 18-row action table with keys and descriptions. This matches anchor 5 — specific examples cover the common cases. | 5 / 5 |
Workflow Clarity | The setup sequence (install → login → connect → poll state → search → run) is clear, with a well-specified state decision tree (READY / BUILDING / CLIENT_ACTION_REQUIRED / CONFIGURATION_ERROR) including re-polling after user action. The gap: 'skip to **Step 2**' references a step label that never exists in the document (only '#### 1b' appears), leaving the sequence with a dangling branch point. Anchor 4 ('clear sequence with most checkpoints; minor gaps') is the best fit. | 4 / 5 |
Progressive Disclosure | No bundle files exist, and the single SKILL.md is well-sectioned with clear headings and appropriately placed getting-started content. However, at ~170 lines it exceeds the under-50-lines simple-skill exception, and bulky material (the 18-row actions table, proxy flag reference, clientAction state details) could live in reference files. Anchor 4 ('good structure; most content appropriately placed; minor organization gaps') fits better than 3, since nothing is buried and navigation is easy. | 4 / 5 |
Total | 17 / 20 Passed |