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 lean, well-structured skill body: terse imperative rules, per-platform constraint lists instead of stereotypes, a banned-patterns list, and a pre-delivery quality gate. The main improvements are an explicit feedback loop when the quality gate fails and a single worked example showing one source adapted across two platforms.
Suggestions
Add one short worked example (source post → X version + LinkedIn version) so the adaptation pattern is concrete rather than rule-based.
Make the Quality Gate a loop: state explicitly that any failing check means rewrite that platform version and re-run the gate before delivering.
If per-platform constraint lists grow, move them to one-level-deep reference files (e.g. references/platforms.md) to keep SKILL.md an overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is almost entirely terse directives ("keep it compressed", "avoid hashtags and generic filler", "do not write fake hyper-casual creator copy") with zero explanation of concepts Claude already knows and no padded sections. The only near-redundancy — rule 5 vs. the Banned Patterns list, and the LinkedIn bullet vs. the banned "What do you think?" — is deliberate reinforcement of the skill's one core rule, so every token still earns its place (anchor 5). | 5 / 5 |
Actionability | For an instruction-only skill the guidance is highly actionable: concrete per-platform constraint lists, a literal banned-phrase list to delete and rewrite, a defined output format, and a quality-gate checklist. It stops short of anchor 5 because there is no worked example (e.g. one source post adapted for X and LinkedIn) that would make the adaptation pattern copy-paste concrete — a minor gap consistent with anchor 4. | 4 / 5 |
Workflow Clarity | A clear three-step sequence (primary version → voice fingerprint → adapt by platform) plus posting order, output format, and a pre-delivery Quality Gate checklist ("no copy is duplicated verbatim across platforms") gives explicit validation. It falls at anchor 4 rather than 5 because the gate lacks an explicit fix-and-recheck feedback loop — nothing says what to do when a quality-gate check fails beyond the implied rewrite. | 4 / 5 |
Progressive Disclosure | No bundle files exist (no references/, scripts/, or assets/), and the ~110-line body is well sectioned with the Related Skills section clearly signaling cross-skill dependencies (brand-voice, content-engine, x-api). Structure is good and nothing is mis-filed, but all platform constraint lists are inlined in SKILL.md with no one-level-deep reference files, which is appropriate at this size yet leaves minor organization headroom — anchor 4. | 4 / 5 |
Total | 17 / 20 Passed |