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.
A strong, highly actionable CLI skill body: every command is executable, workflows are sequenced with a concrete example, and content stays lean. The main gaps are minor — repeated agent-to-agent-mode guidance and an abstract 'run in background and poll' instruction without a concrete polling mechanism.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean — a command table, executable examples, and terse principles — and explains nothing Claude already knows. Minor trimmable redundancy: the --agent-to-agent-mode default is stated three times (top note, dedicated section, Key Principles) and async-by-default twice. Fits 'efficient; minor instances that could be trimmed'. | 4 / 5 |
Actionability | Every section gives copy-paste-ready commands with concrete flags (--chat-id, --model, --timeout, --agent-to-agent-mode), jq pipelines, and a complete end-to-end workflow example covering the common cases. Matches 'fully executable; copy-paste ready; specific examples cover the common cases'. | 5 / 5 |
Workflow Clarity | The investigate-workflow example sequences background ask → chat-id follow-up → artifacts list/get with per-step comments, and sections clearly delineate start/continue. However, 'launch as a background process and poll it for completion' lacks a concrete polling mechanism or completion check, leaving checkpoints implicit — anchor 4, not 5. | 4 / 5 |
Progressive Disclosure | No bundle files exist, and the ~150-line single-file CLI reference is well-organized with clear sections (commands, chat, artifacts, workflows, principles). Structure is appropriate for this size, though it exceeds the under-50-lines simple-skill case and could offload the artifacts output-behavior detail. Anchor 4 fits. | 4 / 5 |
Total | 17 / 20 Passed |