Content
63%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 skill delivers strong, executable command-level guidance with a coherent connection workflow, but a massive inline enumeration of entities and redundant 'Search X' actions bloats the file and should be split into a reference file.
Suggestions
Move the 'Intercom Overview' entity/action enumeration into a separate reference file (e.g. references/entities.md) and keep only a brief summary in SKILL.md.
Collapse the 37 redundant 'Search X' entries into a single note that `membrane action list --intent` searches across all entities.
Fix the dangling 'Step 2' reference by labeling the action-search/run section as 'Step 2' or rewording the 'skip to' instruction.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~190-line 'Intercom Overview' enumeration (every entity plus redundant 'Search X' entries for 37 entities) is a heavily padded section occupying most of the file, even though the instructional sections themselves are lean. | 2 / 5 |
Actionability | Copy-paste-ready commands cover the full workflow (`npm install -g @membranehq/cli@latest`, `membrane connection ensure`, `membrane action list/run`, `membrane request`) plus a flag table, fully executable across common cases. | 5 / 5 |
Workflow Clarity | A clear sequence (install -> auth -> connect -> poll for READY -> search -> run -> proxy) with state-based checkpoints and a poll-after-user-action feedback loop; minor gaps such as the undefined 'Step 2' reference prevent a 5. | 4 / 5 |
Progressive Disclosure | Section structure is reasonable, but the large inline entity/action enumeration clearly belongs in a separate reference file and there are no bundle files or external references to offload it. | 3 / 5 |
Total | 14 / 20 Passed |