Content
75%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 well-structured, lean instruction skill with a clear workflow, concrete per-platform guidance, security handling for untrusted source material, and a closing quality gate. The main gap is the lack of an explicit pre-publish confirmation/validation loop for what is fundamentally a batch, hard-to-reverse outward-facing operation.
Suggestions
Add an explicit pre-publish checkpoint to the workflow (e.g. 'Before posting: confirm target accounts and timing with the user; treat publication as irreversible') to turn the Quality Gate into a true validate→fix→retry feedback loop.
Tighten the platform rule lists by deduplicating the repeated 'keep it concise/readable/direct' cadence into a single shared constraint, then list only what differs per platform.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean throughout — short bullets, no explaining of what X/LinkedIn/Threads are, no library tutorials — and assumes Claude's competence; it is efficient rather than maximally trimmed, with a few list items ('keep it readable and direct', 'preserve the author's cadence') that restate the same idea across platforms. | 4 / 5 |
Actionability | Concrete, specific guidance for an instruction-only skill: explicit tool handoffs ('Run brand-voice first', 'Use content-engine first'), a delete-and-rewrite Banned Patterns list, and per-platform rules like 'lead with the sharpest claim or artifact'; some platform rules remain abstract directives ('keep it compressed') rather than testable criteria, keeping it at 4. | 4 / 5 |
Workflow Clarity | A clear three-step sequence (primary version → voice fingerprint → adapt by platform) plus a terminal 'Quality Gate' that functions as a validation checkpoint before delivery; because this is a batch, outward-facing operation, an explicit pre-publish confirmation/feedback loop would push it to 5, and its absence is the main gap. | 4 / 5 |
Progressive Disclosure | Single-file skill with well-organized, clearly headed sections (When to Activate, Core Rules, Untrusted Source Material, Workflow, Posting Order, Banned Patterns, Output Format, Quality Gate, Related Skills) and no nested references; no bundle files exist to verify, so structure is good but not the exemplar split-file case that defines a 5. | 4 / 5 |
Total | 16 / 20 Passed |