audience
/audience
Podpowiada, jak przekazywać wyniki analizy danych decydentom: tytuły akcji, odpowiedź na początku, niepewność prostymi słowami.
- Autor
- steffen74 ↗
- Repozytorium
- agentic-data-science ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 12.11.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 1
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 12.11.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Podpowiada, jak przekazywać wyniki analizy danych decydentom: tytuły akcji, odpowiedź na początku, niepewność prostymi słowami.
Oryginalny opis (EN)
Rules for communicating data science results to decision-makers and technical readers - action titles, answer first, uncertainty and limits in plain words. Use when writing memos, chart titles, README sections, presentations or model documentation.
Jak działa
- Decision-maker
- Technical reader
- Charts
Podgląd SKILL.md
--- name: audience description: Rules for communicating data science results to decision-makers and technical readers - action titles, answer first, uncertainty and limits in plain words. Use when writing memos, chart titles, README sections, presentations or model documentation. --- # Audience Every result is written for a named reader from canvas block 3. Before writing, state who it is and what they already know. ## Decision-maker - **Answer first.** The first sentence answers the question; details follow. - **Action titles.** Every chart and section title states the conclusion as a sentence (“Rainy mornings need a third fewer bikes in the city centre”), not the topic (“Morning rentals by weather”). - **Few numbers.** At most three per message, rounded, with unit and comparison (vs. baseline, vs. last year, vs. today). - **Uncertainty in plain words.** “In 8 of 10 weeks the forecast was within 15 %” instead of standard errors; say what the uncertainty means for the decision. - **Association, not cause.** “Sales are higher on warm days” unless the analysis was designed to show cause. - **Limits are part of the answer.** Say what the data cannot tell and under which conditions the recommendation does not hold. - **Recommend, with conditions.** What to do, starting when, under which conditions; and where not to rely on the analysis. ## Technical reader - How to reproduce everything with one command. - What was decided and why (link `DECISIONS.md`), especially data decisions and the split. - Results per segment, next to the baseline, with variance across runs where relevant. - Known weaknesses and what you would try next. ## Charts One message per chart. Label axes with units. Order categories meaningfully (weekdays Monday to Sunday, not alphabetically). Show uncertainty (confidence intervals, ranges) when comparisons are made. Prefer direct labels over legends.
Wersje
W bazie nie ma innych skilli o nazwie audience.
Migawka bazy z 13.11.2026 · wersja paczki 857907c628bb (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
Brak podobnych skilli w bazie.