Content
81%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 highly actionable with well-sequenced, validated workflows and real bundle scripts, but it carries redundant config/server content that could be trimmed or externalized for better token efficiency.
Suggestions
Trim or externalize the 'Configuration Reference' and 'Common Email Servers' table, since the body states configuration is pre-configured and the scripts load it automatically.
Move the provider server table into a separate reference file referenced one level deep to improve progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient command documentation, but the 'Common Email Servers' table and 'Configuration Reference' duplicate information already conveyed and partly contradicted by the 'pre-configured' note, adding tokens Claude doesn't need. | 3 / 5 |
Actionability | Fully executable commands with concrete flags, options, and copy-paste-ready examples covering send, HTML, attachments, multiple recipients, and multi-account usage. | 5 / 5 |
Workflow Clarity | Clear sequencing with explicit validation checkpoints: the --confirmed gate and pre-send review for destructive send operations, plus the error-to-check-LobsterAI-Settings feedback loop for failures. | 5 / 5 |
Progressive Disclosure | SKILL.md acts as an overview pointing to real bundle scripts (scripts/imap.js, scripts/smtp.js) one level deep, though the config and server tables are inlined rather than split into separate reference files. | 4 / 5 |
Total | 17 / 20 Passed |