interviewing
/interviewing
Ustala z osobą decyzyjną, czego naprawdę chce, zadając jedno pytanie naraz w rozmowie, zanim zapadnie decyzja o pracy lub planie.
- Autor
- talgent-work ↗
- Repozytorium
- workbench ↗
- Licencja
- Apache-2.0
- Ostatnia zmiana repozytorium
- 13.11.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 1
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 13.11.2026, bez odwołań do wycofanych modeli i API.
Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.
Co robi i kiedy
Ustala z osobą decyzyjną, czego naprawdę chce, zadając jedno pytanie naraz w rozmowie, zanim zapadnie decyzja o pracy lub planie.
Wyzwalacze z opisu
Oryginalny opis (EN)
Find with the person you work for - a Standalone Work's owner, or whom a Work of a Project's product context names - what they actually want, one question at a time in conversation. Use whenever something needs their decision - what the Work should produce, a plan, a change to either - and when another Agent needs their answer.
Jak działa
- The way
- Asking
- Arriving
- For another Agent
Podgląd SKILL.md
--- name: interviewing description: Find with the person you work for - a Standalone Work's owner, or whom a Work of a Project's product context names - what they actually want, one question at a time in conversation. Use whenever something needs their decision - what the Work should produce, a plan, a change to either - and when another Agent needs their answer. --- # Interviewing The person rarely holds the answer ready: they see where they want to go only dimly, and fog lies between. Asking them to fill in what the scenario must settle gets an empty phrase or an answer made up on the spot. Interviewing is wayfinding instead: finding the way there together, one question at a time. The decisions are theirs and finding facts is yours, so you never answer their side for them. ## The way Keep in mind, as the conversation goes: - **Destination**: what will be different when the work is done, in the person's words. - **Decided**: what the person has settled so far. - **Fog**: what you can tell is coming but cannot yet ask precisely. - **Out of scope**: what the person ruled beyond the destination. 1. **Destination first.** Ask what they would point at, once it is done, to say it worked. It fixes the scope, so every later question hangs on it. 2. **Breadth before depth.** Sketch the open questions across the whole way before going deep on any one, so one detail does not swallow the interview. 3. **The frontier.** Ask only a question you can state precisely now and that needs no answer you have not heard. One you cannot yet state precisely stays in the fog until an answer clears it. Take the one that clears the most. ## Asking 1. Look up the facts the next question needs, in what the Work holds: the files and repositories, and in a Work of a Project what its product context names. Facts that only later questions need wait for those questions, so the person is not kept waiting on research their answer may make moot. 2. Ask from where the person stands: a concrete situation, an example, a consequence, a trade-off, what would make them call the work a failure. When they cannot say, show a rough take to react to: a sample, an outline, a draft criterion. 3. Write that one question as your message, in the question format below, and end your Turn there. They answer in their own words, so they can adjust or overrule rather than only pick an option you offered. 4. Read the reply: what it decides, what fog it clears, what it rules out, and the questions it opens. Then ask the next. ### Question format Write in the person's language, with these three blocks separated by blank lines. Use one emoji and a bold heading per block; keep each explanation to one or two sentences. > 💬 **Question: <the concrete decision to make>** > Explain the context and what this choice affects. > > 💡 **Recommended: <option>** > Explain why it fits the person's goals and constraints, including its main trade-off. > > 🔀 **Alternative: <option>** > Explain how it differs, when it would fit better, and why it is less suitable for the current situation. Include only meaningful alternatives, repeating the alternative block when needed. Omit it when none is useful; for an open question with no grounded recommendation yet, use only the question block. Keep the choice open to the person's own answer. ## Arriving The pull to start the work is the sign you have reached the edge of what is decided. The way is clear when no fog remains toward the destination. Write what the scenario must settle from what was decided, with what was ruled out as its limits, in the person's words where they gave them. A part no answer supports is still fog: ask about it. Done when every part of what the scenario must settle rests on the person's answers. ## For another Agent When another Agent asks a question only the person you work for can answer, ask them in conversation, then answer it where it asked, with their answer.
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · talgent-work/workbench | 100 | Apache-2.0 |
| remotivated/job-hunt-skills | 100 | MIT |
| avataristvan/briefing-interviewer | 100 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki bce964895a56 (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
- 100brainstorming
Ustala, czego ktoś naprawdę chce i dlaczego, przed tworzeniem funkcji, komponentu lub planu, zadając jedno pytanie naraz. - 100brainstorming
Ustala, czego ktoś naprawdę chce i dlaczego, przed tworzeniem funkcji, komponentu lub planu, zadając jedno pytanie naraz. - 100brainstorming
Ustala, czego ktoś naprawdę chce i dlaczego, przed tworzeniem funkcji, komponentu lub planu, zadając jedno pytanie naraz. - 100brainstorming
Ustala, czego ktoś naprawdę chce i dlaczego, przed tworzeniem funkcji, komponentu lub planu, zadając jedno pytanie naraz. - 100brainstorming
Ustala, czego ktoś naprawdę chce i dlaczego, przed tworzeniem funkcji, komponentu lub planu, zadając jedno pytanie naraz. - 100brainstorming
Ustala, czego ktoś naprawdę chce i dlaczego, przed tworzeniem funkcji, komponentu lub planu, zadając jedno pytanie naraz.