Content
88%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, highly actionable playbook with clear sequenced workflows, built-in feedback loops, and a properly signaled one-level-deep reference file. The two minor weaknesses are token padding (a Keywords section and re-explanations of the standard ADKAR definitions) and slight duplication between the body's resistance section and the reference file.
Suggestions
Remove the '## Keywords' section — it duplicates trigger terms already present in the frontmatter description and adds no guidance value.
Drop or compress the 'What it is:' definition lines for each ADKAR phase (Claude already knows the standard model) and lead with the mistake/shortcut guidance that is the skill's real value.
Trim the body's resistance-patterns table to a pointer into `references/change-playbook.md` (which has full counter-strategies) to reduce duplication between SKILL.md and the reference.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and prescriptive with short, punchy bullets (e.g., "Declaring victory at go-live. Changes fail because they're never reinforced."), but the "## Keywords" section duplicates the frontmatter description and the "What it is:" lines re-state standard ADKAR definitions Claude already knows — minor padding that could be trimmed. | 4 / 5 |
Actionability | For an instruction-only skill the guidance is fully concrete and executable: week-by-week numbered communication sequences ("1. Week -2: Announce the why + go-live date"), a resistance-pattern-to-response table, a knowledge-transfer method table, and verbatim phrasing rules ("Never: 'We can't share the reasons.'"). | 5 / 5 |
Workflow Clarity | The change-type playbooks are clearly sequenced with explicit feedback loops built in ("Week 2: Adoption check-in (who's using it? Who isn't?)", "Month 2: First retrospective — what's working, what needs adjustment") and the "Key Questions" and "Red Flags" sections act as checklists. This is not a destructive or batch operation, so no validation cap applies. | 5 / 5 |
Progressive Disclosure | There is a single, well-signaled, one-level-deep reference (`references/change-playbook.md`) whose listed contents match the actual file. However, the SKILL.md body is a full playbook rather than an overview, and the resistance-patterns table duplicates material that also appears in the reference file, so content splitting is good but not optimal. | 4 / 5 |
Total | 18 / 20 Passed |