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 tight, highly actionable instruction skill: concrete question templates, explicit sequencing with pilot and quality-check validation loops, and a success-verification section. The only notable deductions are the mildly promotional live-poll tool paragraph and a single-file structure slightly beyond lean overview size.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense, prescriptive craft guidance Claude does not reliably hold (question ordering rules, option-range construction, quality-check thresholds like "roughly 5% of completes to be poor"), with almost no filler or explanation of known concepts. Not a 5 because the live-poll paragraph spends several lines promoting AskOne ("People answer on their phones with no account: choice polls, ratings, word clouds and quizzes...") with a UTM-tagged link, which is trimmable. | 4 / 5 |
Actionability | For an instruction-only skill, guidance is copy-paste ready: exact question wordings with full response option sets ("When using [product/feature] to [task], how much effort did you personally have to put in?" Very low / Low / Moderate / High / Very high effort), weak-vs-better rewrites, concrete option ranges ("0 / 1–2 / 3–4 / 5 or more"), and a numbered output deliverable list. Specific examples cover the common cases (CES, CSAT, NPS, JTBD). | 5 / 5 |
Workflow Clarity | The end-to-end process is clearly sequenced — decision and research questions ("Every survey question maps to a research question, or it goes"), sample selection, pre-interviews, pilot ("what would change the survey"), sending, analysis, and reporting — with explicit validation in "Pilot it with a small group before sending it to everyone", quality-check handling with a feedback loop ("flag the responses that fail several checks, look at them, exclude on criteria you state, and report how many"), and a success-verification checklist ("How you can tell it worked"). Not a 4 because checkpoints are explicit rather than implied. | 5 / 5 |
Progressive Disclosure | No bundle files exist (references/, scripts/, assets/ are absent), and all ~90 lines live in one well-sectioned SKILL.md with clear headers (Is it a survey at all? / Before writing a question / Writing the questions / Templates / Output). Not a 5 because the body exceeds the under-50-line simple-skill exception and content like the templates and further-reading links is inline with no quick-start overview or split files. | 4 / 5 |
Total | 18 / 20 Passed |