Content
36%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 body is a catalogue advertisement for an upstream bundle rather than a usable skill: it repeats the description, provides a single low-value command, and delegates all real guidance to an external URL. Structure and brevity are fine, but actionability and workflow clarity are weak because no executable steps for obtaining or using the skill's content are given.
Suggestions
Replace the `open <url>` snippet with the concrete commands to fetch and install the bundle (e.g., a git clone command and the target skills-directory path), so the install step is executable.
Inline the core actionable guidance (e.g., a paywall/upsell design checklist or the key conversion patterns) instead of deferring 100% of the content to the external repo, or ship the relevant files in references/ and link to them locally.
Remove the verbatim repetition of the description in 'What it does' and the attribution blockquote, or replace them with content that adds information the agent does not already have.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and mostly efficient, but it repeats the frontmatter description verbatim in 'What it does' ("Design and optimize upgrade screens, paywalls, and upsell modals. Useful for SaaS conversion design and pricing-page experiments.") and includes filler like '> Curated from Corey Haines.' — unnecessary explanation that could be trimmed, matching the 'mostly efficient but includes some unnecessary explanation' anchor. It is not score 2 because there is no multi-paragraph padding or explanation of concepts Claude already knows. | 3 / 5 |
Actionability | The only concrete command is `open https://github.com/coreyhaines31/marketingskills`, which merely opens a browser; the instruction to 'install the upstream bundle into your active agent's skills directory' gives no specific steps (no clone command, no target path, no install mechanism). This matches anchor 2 ('minimal concrete guidance; high-level hints but missing the specific steps to execute'). It avoids a 1 because one concrete, executable command and a specific URL are present. | 2 / 5 |
Workflow Clarity | A rough sequence exists (discover entry → install upstream bundle → invoke by name), but the critical middle step is undefined — how to actually obtain and install the bundle is delegated to 'Inspect the upstream README for exact paths'. This matches anchor 2 ('rough sequence present but many gaps; steps poorly defined'). No validation checkpoints exist, though the skill is not destructive or batch-oriented, so no additional cap applies. | 2 / 5 |
Progressive Disclosure | The body has clean section structure (What it does / Source / How to use) and no nested references, but the SKILL.md is a stub: it contains no substantive guidance and defers everything to an external GitHub URL rather than to local bundle files (no references/, scripts/, or assets/ exist). The pointer to the upstream repo is signaled but the actual detail is one level away on an external site with no guarantee of the promised structure, fitting anchor 3 ('some structure but could be better organized; references present but not clearly signaled') better than the well-split anchor 5. | 3 / 5 |
Total | 10 / 20 Passed |