Skip to content
Skills Library
Management & product · Planning

dev-deep-interview

/dev-deep-interview

Socratic questioning to crystallize vague requirements into a testable spec. Use when input is ambiguous, when the user asks for "deep interview", or before dev-autopilot if the brief is too broad. Pairs with @echo-analyst.

Repository
evo-nexus ↗
Licence
Apache-2.0
Last repository change
12.05.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 12.05.2026, without references to retired models or APIs.

Licence

Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.

What it does and when to use it

Socratic questioning to crystallize vague requirements into a testable spec. Use when input is ambiguous, when the user asks for "deep interview", or before dev-autopilot if the brief is too broad. Pairs with @echo-analyst.

Triggers from the description

deep interview
Original author description

Socratic questioning to crystallize vague requirements into a testable spec. Use when input is ambiguous, when the user asks for "deep interview", or before dev-autopilot if the brief is too broad. Pairs with @echo-analyst.

How it works

  1. Use When
  2. Do Not Use When
  3. Goal
  4. Workflow
  5. Rules
  6. Pairs With
  7. Failure Modes To Avoid
Original SKILL.md headings · 102 lines, 618 words
SKILL.md preview
---
name: dev-deep-interview
description: Socratic questioning to crystallize vague requirements into a testable spec. Use when input is ambiguous, when the user asks for "deep interview", or before dev-autopilot if the brief is too broad. Pairs with @echo-analyst.
---

# Dev Deep Interview

Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.

Deep Interview transforms vague ideas into concrete, testable specifications through Socratic questioning. It is the front gate of any high-stakes work — refusing to proceed until ambiguity is below an acceptable threshold.

## Use When
- User has a vague idea ("build me something cool", "improve performance", "fix the UX")
- Before `dev-autopilot` when input lacks file paths, function names, or concrete anchors
- User explicitly asks for "deep interview", "interview me", "ask me questions first"
- High-stakes work where the cost of misunderstanding is high (auth, billing, migrations)

## Do Not Use When
- User has already provided a detailed spec or plan
- Task is trivially clear (typo fix, single-line change)
- User says "just do it" or "skip the questions"

## Goal
Drive ambiguity below 20% before any code or plan is generated. Output a spec that an executor agent can implement without further clarification.

## Workflow

### Phase 1 — Initial Assessment
Read the user's input. Score it on these dimensions (1-5 each, 5 = clear):
- **Domain clarity**: do you understand WHAT this is about?
- **Scope**: do you know what is in/out of scope?
- **Success criteria**: do you know what "done" looks like?
- **Constraints**: do you know technical/business limits?
- **Stakeholders**: do you know who cares and what they want?

If average ≥ 4: skip interview, proceed.
If average < 4: enter interview loop.

### Phase 2 — Socratic Loop
Ask **one question at a time** using AskUserQuestion (with 2-4 multiple-choice options when possible). Each question must:
- Target the lowest-scoring dimension
- Be answerable in under a sentence (or by clicking an option)
- Eliminate at least one ambiguity

Common question types:
- **Scope**: "Should X be included or out of scope?"
- **Trade-off**: "Optimize for speed or simplicity?"
- **Stakeholder**: "Who is the primary user — admin, customer, internal?"
- **Constraint**: "Any budget/time/tech constraints?"
- **Success**: "How will we know this worked?"

Stop the loop when all dimensions ≥ 4 OR after 8 questions (whichever first — long interviews lose user patience).

### Phase 3 — Spec Output
Write the spec to `workspace/projects/specs/[C]deep-interview-{name}.md` with this structure:

```markdown
# Deep Interview Spec — {topic}

**Date:** {iso}
**Ambiguity score:** {avg}/5

## Context
[1-2 sentences on the problem and why it matters]

## In Scope
- [item 1]
- [item 2]

## Out of Scope
- [explicit non-goals]

## Success Criteria
- [testable criterion 1]
- [testable criterion 2]

## Constraints
- [tech / business / time]

## Open Questions
- [items where ambiguity remains, with risk level]

## Suggested Next Step
- `dev-autopilot` (if all dimensions ≥ 4)
- `dev-plan` (if some dimensions still < 4 but you want to start scoping)
- Manual implementation (if the spec is small enough)
```

## Rules
- **One question at a time.** Never batch.
- **Use clickable options** (AskUserQuestion) whenever possible.
- **Don't ask codebase facts** — spawn `@scout-explorer` to look them up.
- **Stop at 8 questions.** Longer kills user patience.
- **Always output a spec file**, even if interview was short.

## Pairs With
- `@echo-analyst` — for deeper requirements gap analysis after the interview
- `dev-autopilot` — natural next step once spec is ready
- `dev-plan` — if you want to scope further before execution

## Failure Modes To Avoid
- **Interrogation**: 15 questions in a row. Stop at 8.
- **Codebase questions**: "What framework do you use?" → look it up yourself.
- **Vague answers accepted**: "I want it fast" → push: "Faster than what — the current 2s, or the user-perceived 200ms?"
- **No spec output**: interview without persistence is wasted.

Versions

The database has no other skills named dev-deep-interview.

Database snapshot: 13.11.2026 · package version da5512360c58 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 90deep-interview
    Socratic deep interview with mathematical ambiguity gating. Activate when user says: deep interview, interview me, ask me everything, don't assume, make sure you understand, ouroboros, socratic, I have a vague idea, not sure exactly what I want.
  • 100v1-interview-me
    Use when an idea, feature request, or ticket needs structured questioning before planning. Triggers on "interview me", "flesh out this idea", or "help me spec this".
  • 90dev-deep-dive
    Two-stage investigation — causal trace (via @trail-tracer) followed by requirements crystallization (via @echo-analyst + dev-deep-interview). Use when a problem is both unclear in cause AND unclear in scope.
  • 95interview
    Use when the requested deliverable is vague, broad, ambiguous, requirement-light, or missing users, constraints, or acceptance criteria and needs requirements discovery before implementation planning or execution.
  • 100deep-interview
    Use when a vague task needs a clarification interview and structured brief. Triggers — "clarify task", "define requirements".
  • 100requirements-interview
    Conduct deep multi-round requirements interviews to produce executable SPEC, PRD, and PLAN drafts. Use when user says "clarify requirements", "采访需求", "extract requirements", "define scope", "analyze needs", or when requirements are ambiguous and need interactive clarification.…

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.