Content
57%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 action-dense and provides a workable Membrane-based workflow with a well-handled connection state machine, but it is undermined by broken step cross-references, empty stub sections (Azuqua Overview, Popular actions), and a docs link pointing to Workato rather than Azuqua. Organization is serviceable but mediocre for a single-file skill.
Suggestions
Fix the broken workflow references: define the numbered steps (there is a "1b" with no "1a" and a "skip to Step 2" pointing at a nonexistent step), and add a post-run validation step (check the output field / handle errors) to close the feedback loop.
Remove or fill the empty sections: either populate "Azuqua Overview" and "Popular actions" with actual content (e.g., the top actions per connector) or delete them — the stub bullets and "Use action names and parameters as needed" line are pure padding.
Correct the "Official docs" link (https://help.workato.com/ is Workato's documentation, not Azuqua's) and consider moving the Membrane auth/proxy details into a reference file to slim the main body.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The bulk is lean, executable CLI commands, but there is real padding: the "Azuqua Overview" section is bare stub bullets (Flow / Execution / Connector / Datastore) plus the filler line "Use action names and parameters as needed", and "Popular actions" is a header that lists nothing popular. This fits anchor 3 (mostly efficient but includes some unnecessary content that could be tightened) rather than anchor 2, since the padding is limited to a few sections. | 3 / 5 |
Actionability | Concrete, executable commands appear throughout (npm install -g @membranehq/cli@latest, membrane connection ensure "https://azuqua.com/" --json, membrane action run <actionId> --connectionId=... --input '...' --json) with a complete proxy flag table. Anchor 4 (mostly executable, minor gaps) — not 5 because "skip to Step 2" points to a step that doesn't exist and the Popular actions section delivers no actual runnable examples. | 4 / 5 |
Workflow Clarity | A real sequence exists (install → login → connection ensure → poll → search → run → proxy fallback) and the connection state machine (READY / CLIENT_ACTION_REQUIRED with agentInstructions handling / CONFIGURATION_ERROR) is a genuine checkpoint loop, but the numbering is incoherent: "Step 2" is referenced but never defined, "1b" appears without a "1a", and there is no validation of action run results. This is anchor 3 (steps listed but checkpoints broken or implicit), below anchor 4's "clear sequence with most checkpoints present". | 3 / 5 |
Progressive Disclosure | This is a single-file skill with no bundle files (no references/, scripts/, or assets/), clear section headers, and no nested references, but the ~140-line body inlines material that could be split (full Membrane auth walkthrough, proxy flag table) and contains empty stub sections that hurt navigation. Anchor 3 (some structure but could be better organized) rather than anchor 4. | 3 / 5 |
Total | 13 / 20 Passed |