Content
38%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 skill has a reasonable skeleton (pre-flight checks, classification, interaction rules) but its substance is entirely delegated to sub-skill and resource files that are missing from the bundle, and the Instructions section is generic filler. As written, it cannot be executed to produce the specification and phased plan it promises.
Suggestions
Ship the referenced bundle files (sub-skills/*.md, resources/implementation-playbook.md) or inline their essential content — currently every substantive reference is dangling.
Fix the Knowledge Modules list: Tasks (./sub-skills/tasks.md) and Verification (./sub-skills/verification.md) each appear twice (lines 71-74), indicating a copy-paste error.
Replace the generic Instructions and 'Use this skill when' boilerplate with concrete guidance: the actual specification questions per track type and the required plan/task structure.
Complete the workflow after classification: add steps for drafting the specification, building the phased plan, and verifying the output.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and mostly free of concept explanations, but contains boilerplate filler that adds nothing ("Use this skill when: Working on new track tasks or workflows", "The task is unrelated to new track", "Apply relevant best practices and validate outcomes"). This is 'mostly efficient but includes some unnecessary explanation or could be tightened', not the lean 5 or the padded 2. | 3 / 5 |
Actionability | Concrete guidance exists only in pre-flight checks (file paths, the /conductor:setup fallback) and the track-type menu, but the core deliverable — writing the specification and phased plan — has no executable guidance: the Instructions section is high-level hints ("Clarify goals, constraints, and required inputs") and every Knowledge Module defers to files that do not exist in the bundle. This matches 'minimal concrete guidance; high-level hints but missing the specific steps to execute' rather than a 3, because the missing specifics are the skill's actual purpose. | 2 / 5 |
Workflow Clarity | A sequence exists (pre-flight checks with an explicit error branch → context loading → classification → interactive gathering) but the workflow stops after classification: there are no steps for producing the spec or plan, no checkpoints in the gathering phase, and the module list is corrupted (Tasks and Verification each appear twice). Validation exists in pre-flight, so this sits above the 'many gaps, validation absent' anchor of 2 but below the 'most checkpoints present' anchor of 4. | 3 / 5 |
Progressive Disclosure | References are prominently signaled (a numbered list of ./sub-skills/*.md modules and a conditional reference to resources/implementation-playbook.md), but no bundle files exist anywhere — every referenced path is dangling, so the navigation structure points at nothing and the substantive content is simply absent. Additionally the module list contains duplicate entries (Tasks and Verification each listed twice). Broken references make the structure effectively unusable, below the 'references present but not clearly signaled' anchor of 3. | 2 / 5 |
Total | 10 / 20 Passed |