Content
78%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 positioning workflow: concrete search commands, decision criteria, and a complete output template leave Claude knowing exactly what to do at every phase. The main weaknesses are the absence of validation/feedback steps in the workflow and the monolithic single-file layout — the full output template should be split into a reference file.
Suggestions
Move the Phase 4 output-document template into a references/ file (e.g., references/positioning-doc-template.md) and replace it in SKILL.md with a one-line pointer, keeping the main file as a lean overview of the workflow.
Add an explicit validation checkpoint between Phase 2 and Phase 4, e.g., "Before writing the doc: every differentiator in the 2C table must have a proof point; if missing, either find evidence via search or drop the claim" — this creates the missing feedback loop.
Remove the duplicate 'Trigger Phrases' section (it repeats the 'When to Use' list and the frontmatter description) and trim rationale sentences like 'Ad copy reveals what competitors believe converts' to tighten token usage.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is largely lean — it names April Dunford's framework without re-teaching it, gives decision criteria as terse conditionals, and avoids explaining marketing concepts Claude already knows. A couple of sections could be trimmed: the "Trigger Phrases" section duplicates the "When to Use" list and the frontmatter description, and "Ad copy reveals what competitors believe converts" is a rationale sentence that could be cut. Not 5 due to those redundancies; not 3 because there is no concept re-teaching or padding of the kind the 3-anchor example shows. | 4 / 5 |
Actionability | Guidance is fully executable for an instruction-only skill: copy-paste search strings ("Search: '[competitor name]' site:[competitor-url]"), concrete extraction checklists, decision criteria tables, a complete fill-in messaging hierarchy, a full output document template, and an explicit save path ("positioning-[YYYY-MM-DD].md"). Every phase has concrete steps, examples, or a ready template rather than high-level hints, so there are no minor gaps that would drop it to 4. | 5 / 5 |
Workflow Clarity | Phases 0–4 are clearly sequenced (intake → research → framework → map → output), and Phase 2A's decision criteria ("If you'd spend 50% of sales calls explaining the category → don't create a new one") act as explicit checkpoints. However, there are no validation steps for the research output — e.g., verifying that every claimed differentiator has a proof point before writing the doc, or sanity-checking the 2x2 map against extracted competitor data. Not 5 because feedback loops / explicit validation steps are absent (the operation is research-and-writing, not destructive or batch, so the cap at 3 does not apply); not 3 because the sequence is complete and checkpoints are explicit via the decision criteria rather than merely implied. | 4 / 5 |
Progressive Disclosure | The body (~250 lines) is well-organized with clear ## / ### sectioning and no nested references, but everything lives inline in SKILL.md — no references/, scripts/, or assets/ exist. The ~100-line Phase 4 output template is exactly the kind of large, self-contained artifact that belongs in a one-level-deep reference file, so content that should be separate is inline. Not 4 because the inlined full-document template is substantial content that clearly belongs in its own file rather than a minor organization gap; not 2 because the structure is strong throughout — sections are clearly headed and nothing is buried. | 3 / 5 |
Total | 16 / 20 Passed |