Content
46%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 has a genuinely good progressive-disclosure design on paper — a selective-reading content map and a template index — but every referenced file is missing from the bundle, so the navigation leads nowhere, and the actual orchestration workflow exists only as a usage example without validation checkpoints. Two sections (When to Use, Limitations) are vague filler that spend tokens without adding skill-specific value.
Suggestions
Ship the referenced bundle files (project-detection.md, tech-stack.md, agent-coordination.md, scaffolding.md, feature-building.md, and the templates/ tree) — the entire content map currently points at paths that do not exist, which is the single biggest defect.
Replace the filler "When to Use" section with concrete trigger conditions, and make the orchestration steps a real workflow section with explicit checkpoints (e.g., confirm the plan with the user before scaffolding, verify the scaffold builds/starts before handing off).
Fix the "Templates (13)" header — only 12 templates are listed — and delete or rewrite the generic Limitations boilerplate to address this skill's actual risks (e.g., tech-stack choices that conflict with an existing codebase).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is table-driven and mostly lean, but contains unnecessary filler: "This skill is applicable to execute the workflow or actions described in the overview" says nothing, and the Limitations section is generic boilerplate ("Do not treat the output as a substitute for environment-specific validation"). Not 4 because two full sections add tokens without skill-specific value; not 2 because there is no concept re-explanation and the tables are tight. | 3 / 5 |
Actionability | The content-map table gives concrete read-when guidance and the Instagram-clone walkthrough is a concrete illustration, but no executable commands, code, or specific procedures appear — the actual instructions are delegated to files that are not in the bundle. Not 4 because the guidance stops at "read this file" and high-level steps like "Coordinate agents" lack any operational detail; not 2 because the read-when table and worked example go beyond vague high-level hints. | 3 / 5 |
Workflow Clarity | A sequence exists ("1. Project type... 2. Tech stack... 3. Create plan... 4. Coordinate agents... 5. Report progress... 6. Start preview") but only inside a usage example, with no validation or verification checkpoints anywhere (e.g., nothing about confirming the plan with the user or verifying the scaffolded app runs). Not 4 because checkpoints are absent rather than just minor; not 2 because a recognizable 6-step sequence is present rather than a rough fragment. | 3 / 5 |
Progressive Disclosure | The content map and template index are well-designed one-level-deep navigation in form, but none of the referenced files exist in the bundle — no references/, scripts/, or assets/ directories are present, so every link (project-detection.md, tech-stack.md, agent-coordination.md, scaffolding.md, feature-building.md, templates/SKILL.md, and all 12 templates/*/TEMPLATE.md) dangles. Navigation that resolves to nothing is unusable in practice. Not 3 because broken references are a structural failure beyond "could be better organized"; not 1 because the body itself is well-sectioned and the intended structure is clear. | 2 / 5 |
Total | 11 / 20 Passed |