Content
92%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 an exemplary lean CLI skill: executable commands for every action, clear prerequisite permissions, an explicit input-discovery step, and a confirmation checkpoint before the outward-facing send. The only flaw is minor redundancy in the "Ideas to try" section. No bundle files exist, so all guidance lives appropriately in SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence (no explanation of what iMessage is), but "Ideas to try" repeats the exact `imsg chats --limit 10 --json` command already shown in two places and restates the watch action — minor content that could be trimmed, matching the score-4 anchor rather than the every-token-earns-its-place score-5 anchor. | 4 / 5 |
Actionability | All four actions are given as fully executable, copy-paste-ready bash commands with realistic flags and a concrete example recipient (`imsg send --to "+14155551212" --text "hi"`), and the inputs section explains exactly how to obtain the required chatId, covering the common cases. | 5 / 5 |
Workflow Clarity | For this simple single-purpose skill the workflow is unambiguous: "Inputs to collect" sequences discovery (chatId from `imsg chats --limit 10 --json`) before history/watch, and the risky send action has an explicit validation checkpoint ("Confirm recipient + message before sending"), satisfying the simple-skill exception. | 5 / 5 |
Progressive Disclosure | The body is ~50 lines with no external reference files (references/, scripts/, assets/ are absent) and no need for them; the content is split into well-organized sections (Overview, Requirements, Inputs, Actions, Notes), which per the guidelines earns a 5 for short skills without external references. | 5 / 5 |
Total | 19 / 20 Passed |