Deep Interview

Inspired by the Ouroboros project which demonstrated that specification quality is the primary bottleneck in AI-assisted development.

• If no usable mission brief is present yet, start by asking: "What should autoresearch improve or prove for this repo?"
• After the mission is clear, collect an evaluator command. If the user leaves it blank, infer one only when repo evidence is strong; otherwise keep interviewing until an evaluator is explicit enough to launch safely.
• Keep the usual one-question-per-round rule, but treat mission clarity and evaluator clarity as hard readiness gates in addition to the normal ambiguity threshold.
• Once ready, do not bridge into omc-plan, autopilot, ralph, or team. Instead run:
  • omc autoresearch --mission "<mission>" --eval "<evaluator>" [--keep-policy <policy>] [--slug <slug>]
• This direct handoff is expected to detach into the real autoresearch runtime tmux session. After a successful handoff, announce the launched session and end the interview lane.

Phase 1: Initialize

1. Parse the user's idea from {{ARGUMENTS}}
2. Detect brownfield vs greenfield:
   • Run explore agent (haiku): check if cwd has existing source code, package files, or git history
   • If source files exist AND the user's idea references modifying/extending something: brownfield
   • Otherwise: greenfield
3. For brownfield: Run explore agent to map relevant codebase areas, store as codebase_context
4. Initialize state via state_write(mode="deep-interview"):

{
  "active": true,
  "current_phase": "deep-interview",
  "state": {
    "interview_id": "<uuid>",
    "type": "greenfield|brownfield",
    "initial_idea": "<user input>",
    "rounds": [],
    "current_ambiguity": 1.0,
    "threshold": 0.2,
    "codebase_context": null,
    "challenge_modes_used": [],
    "ontology_snapshots": []
  }
}

5. Announce the interview to the user:

Starting deep interview. I'll ask targeted questions to understand your idea thoroughly before building anything. After each answer, I'll show your clarity score. We'll proceed to execution once ambiguity drops below 20%.

Your idea: "{initial_idea}" Project type: {greenfield|brownfield} Current ambiguity: 100% (we haven't started yet)

Phase 2: Interview Loop

Repeat until ambiguity ≤ threshold OR user exits early:

Step 2a: Generate Next Question

Build the question generation prompt with:

• The user's original idea
• All prior Q&A rounds (conversation history)
• Current clarity scores per dimension (which is weakest?)
• Challenge agent mode (if activated -- see Phase 3)
• Brownfield codebase context (if applicable)

Question targeting strategy:

• Identify the dimension with the LOWEST clarity score
• Generate a question that specifically improves that dimension
• Questions should expose ASSUMPTIONS, not gather feature lists

Question styles by dimension:

Step 2b: Ask the Question

Use AskUserQuestion with the generated question. Present it clearly with the current ambiguity context:

Round {n} | Targeting: {weakest_dimension} | Ambiguity: {score}%

{question}

Options should include contextually relevant choices plus free-text.

Step 2c: Score Ambiguity

After receiving the user's answer, score clarity across all dimensions.

Scoring prompt (use opus model, temperature 0.1 for consistency):

Given the following interview transcript for a {greenfield|brownfield} project, score clarity on each dimension from 0.0 to 1.0:

Original idea: {idea}

Transcript:
{all rounds Q&A}

Score each dimension:
1. Goal Clarity (0.0-1.0): Is the primary objective unambiguous? Can you state it in one sentence without qualifiers? Can you name the key entities (nouns) and their relationships (verbs) without ambiguity?
2. Constraint Clarity (0.0-1.0): Are the boundaries, limitations, and non-goals clear?
3. Success Criteria Clarity (0.0-1.0): Could you write a test that verifies success? Are acceptance criteria concrete?
{4. Context Clarity (0.0-1.0): [brownfield only] Do we understand the existing system well enough to modify it safely? Do the identified entities map cleanly to existing codebase structures?}

For each dimension provide:
- score: float (0.0-1.0)
- justification: one sentence explaining the score
- gap: what's still unclear (if score < 0.9)

5. Ontology Extraction: Identify all key entities (nouns) discussed in the transcript.

{If round > 1, inject: "Previous round's entities: {prior_entities_json from state.ontology_snapshots[-1]}. REUSE these entity names where the concept is the same. Only introduce new names for genuinely new concepts."}

For each entity provide:
- name: string (the entity name, e.g., "User", "Order", "PaymentMethod")
- type: string (e.g., "core domain", "supporting", "external system")
- fields: string[] (key attributes mentioned)
- relationships: string[] (e.g., "User has many Orders")

Respond as JSON. Include an additional "ontology" key containing the entities array alongside the dimension scores.

Calculate ambiguity:

Greenfield: ambiguity = 1 - (goal × 0.40 + constraints × 0.30 + criteria × 0.30) Brownfield: ambiguity = 1 - (goal × 0.35 + constraints × 0.25 + criteria × 0.25 + context × 0.15)

Calculate ontology stability:

Round 1 special case: For the first round, skip stability comparison. All entities are "new". Set stability_ratio = N/A. If any round produces zero entities, set stability_ratio = N/A (avoids division by zero).

For rounds 2+, compare with the previous round's entity list:

• stable_entities: entities present in both rounds with the same name
• changed_entities: entities with different names but the same type AND >50% field overlap (treated as renamed, not new+removed)
• new_entities: entities in this round not matched by name or fuzzy-match to any previous entity
• removed_entities: entities in the previous round not matched to any current entity
• stability_ratio: (stable + changed) / total_entities (0.0 to 1.0, where 1.0 = fully converged)

This formula counts renamed entities (changed) toward stability. Renamed entities indicate the concept persists even if the name shifted — this is convergence, not instability. Two entities with different names but the same type and >50% field overlap should be classified as "changed" (renamed), not as one removed and one added.

Show your work: Before reporting stability numbers, briefly list which entities were matched (by name or fuzzy) and which are new/removed. This lets the user sanity-check the matching.

Store the ontology snapshot (entities + stability_ratio + matching_reasoning) in state.ontology_snapshots[].

Step 2d: Report Progress

After scoring, show the user their progress:

Round {n} complete.

| Dimension | Score | Weight | Weighted | Gap |
|-----------|-------|--------|----------|-----|
| Goal | {s} | {w} | {s*w} | {gap or "Clear"} |
| Constraints | {s} | {w} | {s*w} | {gap or "Clear"} |
| Success Criteria | {s} | {w} | {s*w} | {gap or "Clear"} |
| Context (brownfield) | {s} | {w} | {s*w} | {gap or "Clear"} |
| **Ambiguity** | | | **{score}%** | |

**Ontology:** {entity_count} entities | Stability: {stability_ratio} | New: {new} | Changed: {changed} | Stable: {stable}

{score <= threshold ? "Clarity threshold met! Ready to proceed." : "Focusing next question on: {weakest_dimension}"}

Step 2e: Update State

Update interview state with the new round and scores via state_write.

Step 2f: Check Soft Limits

• Round 3+: Allow early exit if user says "enough", "let's go", "build it"
• Round 10: Show soft warning: "We're at 10 rounds. Current ambiguity: {score}%. Continue or proceed with current clarity?"
• Round 20: Hard cap: "Maximum interview rounds reached. Proceeding with current clarity level ({score}%)."

Phase 3: Challenge Agents

At specific round thresholds, shift the questioning perspective:

Round 4+: Contrarian Mode

Inject into the question generation prompt:

You are now in CONTRARIAN mode. Your next question should challenge the user's core assumption. Ask "What if the opposite were true?" or "What if this constraint doesn't actually exist?" The goal is to test whether the user's framing is correct or just habitual.

Round 6+: Simplifier Mode

Inject into the question generation prompt:

You are now in SIMPLIFIER mode. Your next question should probe whether complexity can be removed. Ask "What's the simplest version that would still be valuable?" or "Which of these constraints are actually necessary vs. assumed?" The goal is to find the minimal viable specification.

Round 8+: Ontologist Mode (if ambiguity still > 0.3)

Inject into the question generation prompt:

You are now in ONTOLOGIST mode. The ambiguity is still high after 8 rounds, suggesting we may be addressing symptoms rather than the core problem. The tracked entities so far are: {current_entities_summary from latest ontology snapshot}. Ask "What IS this, really?" or "Looking at these entities, which one is the CORE concept and which are just supporting?" The goal is to find the essence by examining the ontology.

Challenge modes are used ONCE each, then return to normal Socratic questioning. Track which modes have been used in state.

Phase 4: Crystallize Spec

When ambiguity ≤ threshold (or hard cap / early exit):

1. Generate the specification using opus model with the full interview transcript
2. Write to file: .omc/specs/deep-interview-{slug}.md

Spec structure:

# Deep Interview Spec: {title}

## Metadata
- Interview ID: {uuid}
- Rounds: {count}
- Final Ambiguity Score: {score}%
- Type: greenfield | brownfield
- Generated: {timestamp}
- Threshold: {threshold}
- Status: {PASSED | BELOW_THRESHOLD_EARLY_EXIT}

## Clarity Breakdown
| Dimension | Score | Weight | Weighted |
|-----------|-------|--------|----------|
| Goal Clarity | {s} | {w} | {s*w} |
| Constraint Clarity | {s} | {w} | {s*w} |
| Success Criteria | {s} | {w} | {s*w} |
| Context Clarity | {s} | {w} | {s*w} |
| **Total Clarity** | | | **{total}** |
| **Ambiguity** | | | **{1-total}** |

## Goal
{crystal-clear goal statement derived from interview}

## Constraints
- {constraint 1}
- {constraint 2}
- ...

## Non-Goals
- {explicitly excluded scope 1}
- {explicitly excluded scope 2}

## Acceptance Criteria
- [ ] {testable criterion 1}
- [ ] {testable criterion 2}
- [ ] {testable criterion 3}
- ...

## Assumptions Exposed & Resolved
| Assumption | Challenge | Resolution |
|------------|-----------|------------|
| {assumption} | {how it was questioned} | {what was decided} |

## Technical Context
{brownfield: relevant codebase findings from explore agent}
{greenfield: technology choices and constraints}

## Ontology (Key Entities)
{Fill from the FINAL round's ontology extraction, not just crystallization-time generation}

| Entity | Type | Fields | Relationships |
|--------|------|--------|---------------|
| {entity.name} | {entity.type} | {entity.fields} | {entity.relationships} |

## Ontology Convergence
{Show how entities stabilized across interview rounds using data from ontology_snapshots in state}

| Round | Entity Count | New | Changed | Stable | Stability Ratio |
|-------|-------------|-----|---------|--------|----------------|
| 1 | {n} | {n} | - | - | - |
| 2 | {n} | {new} | {changed} | {stable} | {ratio}% |
| ... | ... | ... | ... | ... | ... |
| {final} | {n} | {new} | {changed} | {stable} | {ratio}% |

## Interview Transcript
<details>
<summary>Full Q&A ({n} rounds)</summary>

### Round 1
**Q:** {question}
**A:** {answer}
**Ambiguity:** {score}% (Goal: {g}, Constraints: {c}, Criteria: {cr})

...
</details>

Phase 5: Execution Bridge

Autoresearch override: if --autoresearch is active, skip the standard execution options below. The only valid bridge is the direct omc autoresearch --mission ... --eval ... handoff described above.

After the spec is written, present execution options via AskUserQuestion:

Question: "Your spec is ready (ambiguity: {score}%). How would you like to proceed?"

Options:

1. Ralplan → Autopilot (Recommended)

   • Description: "3-stage pipeline: consensus-refine this spec with Planner/Architect/Critic, then execute with full autopilot. Maximum quality."
   • Action: Invoke Skill("oh-my-claudecode:omc-plan") with --consensus --direct flags and the spec file path as context. The --direct flag skips the omc-plan skill's interview phase (the deep interview already gathered requirements), while --consensus triggers the Planner/Architect/Critic loop. When consensus completes and produces a plan in .omc/plans/, invoke Skill("oh-my-claudecode:autopilot") with the consensus plan as Phase 0+1 output — autopilot skips both Expansion and Planning, starting directly at Phase 2 (Execution).
   • Pipeline: deep-interview spec → omc-plan --consensus --direct → autopilot execution

2. Execute with autopilot (skip ralplan)

   • Description: "Full autonomous pipeline — planning, parallel implementation, QA, validation. Faster but without consensus refinement."
   • Action: Invoke Skill("oh-my-claudecode:autopilot") with the spec file path as context. The spec replaces autopilot's Phase 0 — autopilot starts at Phase 1 (Planning).

3. Execute with ralph

   • Description: "Persistence loop with architect verification — keeps working until all acceptance criteria pass"
   • Action: Invoke Skill("oh-my-claudecode:ralph") with the spec file path as the task definition.

4. Execute with team

   • Description: "N coordinated parallel agents — fastest execution for large specs"
   • Action: Invoke Skill("oh-my-claudecode:team") with the spec file path as the shared plan.

5. Refine further

   • Description: "Continue interviewing to improve clarity (current: {score}%)"
   • Action: Return to Phase 2 interview loop.

IMPORTANT: On execution selection, MUST invoke the chosen skill via Skill(). Do NOT implement directly. The deep-interview agent is a requirements agent, not an execution agent.

The 3-Stage Pipeline (Recommended Path)

Stage 1: Deep Interview          Stage 2: Ralplan                Stage 3: Autopilot
┌─────────────────────┐    ┌───────────────────────────┐    ┌──────────────────────┐
│ Socratic Q&A        │    │ Planner creates plan      │    │ Phase 2: Execution   │
│ Ambiguity scoring   │───>│ Architect reviews         │───>│ Phase 3: QA cycling  │
│ Challenge agents    │    │ Critic validates          │    │ Phase 4: Validation  │
│ Spec crystallization│    │ Loop until consensus      │    │ Phase 5: Cleanup     │
│ Gate: ≤20% ambiguity│    │ ADR + RALPLAN-DR summary  │    │                      │
└─────────────────────┘    └───────────────────────────┘    └──────────────────────┘
Output: spec.md            Output: consensus-plan.md        Output: working code

Why 3 stages? Each stage provides a different quality gate:

1. Deep Interview gates on clarity — does the user know what they want?
2. Ralplan gates on feasibility — is the approach architecturally sound?
3. Autopilot gates on correctness — does the code work and pass review?

Skipping any stage is possible but reduces quality assurance:

• Skip Stage 1 → autopilot may build the wrong thing (vague requirements)
• Skip Stage 2 → autopilot may plan poorly (no Architect/Critic challenge)
• Skip Stage 3 → no execution (just a refined plan)

Question: "I see your project uses JWT authentication with passport.js in src/auth/. For this new feature, should we extend the existing auth middleware or create a separate authentication flow?"

Why good: Explored first, then asked an informed question. Never asks the user what the code already reveals.
</Good>

<Good>
Contrarian mode activation:

Round 5 | Contrarian Mode | Ambiguity: 42%

You've said this needs to support 10,000 concurrent users. What if it only needed to handle 100? Would the architecture change fundamentally, or is the 10K number an assumption rather than a measured requirement?

Why good: Challenges a specific assumption (scale requirement) that could dramatically simplify the solution.
</Good>

<Good>
Early exit with warning:

User: "That's enough, just build it" System: "Current ambiguity is 35% (threshold: 20%). Areas still unclear:

• Success Criteria: 0.5 (How do we verify the search ranking algorithm works correctly?)
• Constraints: 0.6 (No performance targets defined yet)

Proceeding may require rework. Continue anyway?" [Yes, proceed] [Ask 2-3 more questions] [Cancel]

Why good: Respects user's desire to stop but transparently shows the risk.
</Good>

<Good>
Ontology convergence tracking:

Round 3 entities: User, Task, Project (stability: N/A → 67%) Round 4 entities: User, Task, Project, Tag (stability: 75% — 3 stable, 1 new) Round 5 entities: User, Task, Project, Tag (stability: 100% — all 4 stable)

"Ontology has converged — the same 4 entities appeared in 2 consecutive rounds with no changes. The domain model is stable."

Why good: Shows entity tracking across rounds with visible convergence. Stability ratio increases as the domain model solidifies, giving mathematical evidence that the interview is converging on a stable understanding.
</Good>

<Bad>
Batching multiple questions:

"What's the target audience? And what tech stack? And how should auth work? Also, what's the deployment target?"

Why bad: Four questions at once — causes shallow answers and makes scoring inaccurate.
</Bad>

<Bad>
Asking about codebase facts:

"What database does your project use?"

Why bad: Should have spawned explore agent to find this. Never ask the user what the code already tells you.
</Bad>

<Bad>
Proceeding despite high ambiguity:

"Ambiguity is at 45% but we've done 5 rounds, so let's start building."

Why bad: 45% ambiguity means nearly half the requirements are unclear. The mathematical gate exists to prevent exactly this.
</Bad>
</Examples>

<Escalation_And_Stop_Conditions>
- **Hard cap at 20 rounds**: Proceed with whatever clarity exists, noting the risk
- **Soft warning at 10 rounds**: Offer to continue or proceed
- **Early exit (round 3+)**: Allow with warning if ambiguity > threshold
- **User says "stop", "cancel", "abort"**: Stop immediately, save state for resume
- **Ambiguity stalls** (same score +-0.05 for 3 rounds): Activate Ontologist mode to reframe
- **All dimensions at 0.9+**: Skip to spec generation even if not at round minimum
- **Codebase exploration fails**: Proceed as greenfield, note the limitation
</Escalation_And_Stop_Conditions>

<Final_Checklist>
- [ ] Interview completed (ambiguity ≤ threshold OR user chose early exit)
- [ ] Ambiguity score displayed after every round
- [ ] Challenge agents activated at correct thresholds (round 4, 6, 8)
- [ ] Spec file written to `.omc/specs/deep-interview-{slug}.md`
- [ ] Spec includes: goal, constraints, acceptance criteria, clarity breakdown, transcript
- [ ] Execution bridge presented via AskUserQuestion
- [ ] Selected execution mode invoked via Skill() (never direct implementation)
- [ ] If 3-stage pipeline selected: omc-plan --consensus --direct invoked, then autopilot with consensus plan
- [ ] State cleaned up after execution handoff
- [ ] Per-round ambiguity report includes Ontology row with entity count and stability ratio
- [ ] Spec includes Ontology (Key Entities) table and Ontology Convergence section
</Final_Checklist>

<Advanced>
## Configuration

Optional settings in `.claude/settings.json`:

```json
{
  "omc": {
    "deepInterview": {
      "ambiguityThreshold": 0.2,
      "maxRounds": 20,
      "softWarningRounds": 10,
      "minRoundsBeforeExit": 3,
      "enableChallengeAgents": true,
      "autoExecuteOnComplete": false,
      "defaultExecutionMode": "autopilot",
      "scoringModel": "opus"
    }
  }
}

Resume

If interrupted, run /deep-interview again. The skill reads state from .omc/state/deep-interview-state.json and resumes from the last completed round.

Integration with Autopilot

When autopilot receives a vague input (no file paths, function names, or concrete anchors), it can redirect to deep-interview:

User: "autopilot build me a thing"
Autopilot: "Your request is quite open-ended. Would you like to run a deep interview first to clarify requirements?"
  [Yes, interview first] [No, expand directly]

If the user chooses interview, autopilot invokes /deep-interview. When the interview completes and the user selects "Execute with autopilot", the spec becomes Phase 0 output and autopilot continues from Phase 1 (Planning).

The 3-Stage Pipeline: deep-interview → ralplan → autopilot

The recommended execution path chains three quality gates:

/deep-interview "vague idea"
  → Socratic Q&A until ambiguity ≤ 20%
  → Spec written to .omc/specs/deep-interview-{slug}.md
  → User selects "Ralplan → Autopilot"
  → /omc-plan --consensus --direct (spec as input, skip interview)
    → Planner creates implementation plan from spec
    → Architect reviews for architectural soundness
    → Critic validates quality and testability
    → Loop until consensus (max 5 iterations)
    → Consensus plan written to .omc/plans/
  → /autopilot (plan as input, skip Phase 0+1)
    → Phase 2: Parallel execution via Ralph + Ultrawork
    → Phase 3: QA cycling until tests pass
    → Phase 4: Multi-perspective validation
    → Phase 5: Cleanup

The omc-plan skill receives the spec with --consensus --direct flags because the deep interview already did the requirements gathering. The --direct flag (supported by the omc-plan skill, which ralplan aliases) skips the interview phase and goes straight to Planner → Architect → Critic consensus. The consensus plan includes:

• RALPLAN-DR summary (Principles, Decision Drivers, Options)
• ADR (Decision, Drivers, Alternatives, Why chosen, Consequences)
• Testable acceptance criteria (inherited from deep-interview spec)
• Implementation steps with file references

Autopilot receives the ralplan consensus plan and skips both Phase 0 (Expansion) and Phase 1 (Planning) since ralplan already produced a Critic-approved plan. Autopilot starts directly at Phase 2 (Execution).

Integration with Ralplan Gate

The ralplan pre-execution gate already redirects vague prompts to planning. Deep interview can serve as an alternative redirect target for prompts that are too vague even for ralplan:

Vague prompt → ralplan gate → deep-interview (if extremely vague) → ralplan (with clear spec) → autopilot

Brownfield vs Greenfield Weights

Brownfield adds Context Clarity because modifying existing code safely requires understanding the system being changed.

Challenge Agent Modes

Each mode is used exactly once, then normal Socratic questioning resumes. Modes are tracked in state to prevent repetition.

Ambiguity Score Interpretation

Task: {{ARGUMENTS}}