research-interview

Quick entry point for structured knowledge elicitation with epistemic tracking. PROACTIVELY activate for: (1) interviews and requirements gathering, (2) problem definition, (3) domain knowledge captur

SKILL.md

GitHub

Research Interview

Structured knowledge elicitation with epistemic tracking for problem definition and requirements gathering.

When to Use

Use this skill when you need to:

  • Elicit knowledge from stakeholders or domain experts
  • Define a problem space systematically
  • Gather requirements with confidence tracking
  • Surface hidden assumptions and constraints
  • Capture domain knowledge for later research

Workflow

Invoke the research-interviewer skill for: "$ARGUMENTS"

Default Parameters

Parameter Value Rationale
output_format PROBLEM-STATEMENT Default; produces actionable problem definition
confidence_threshold 0.85 High confidence required for conclusions
validation_mode balanced Mix of challenging and supportive questioning
epistemic_tracking enabled Track certainty levels throughout

Output Formats

Select based on your goal:

  • PROBLEM-STATEMENT - Clear problem definition with constraints and success criteria
  • REQUIREMENTS - Structured requirements with priority and confidence levels
  • KNOWLEDGE-CORPUS - Domain knowledge capture for research continuation

Interview Techniques Applied

  1. Open-ended exploration - Surface the problem space
  2. Assumption challenging - Identify hidden constraints
  3. Edge case probing - Find boundary conditions
  4. Confidence calibration - Assess certainty levels
  5. Gap identification - Find missing information

Epistemic Labels

All findings are tagged with confidence:

  • VERIFIED - Confirmed by multiple sources or evidence
  • STATED - Reported but not independently verified
  • INFERRED - Derived from other information
  • SPECULATIVE - Hypothesis requiring validation
  • UNKNOWN - Explicit knowledge gap

Output Format

The research interview produces structured output:

<interview-output>
  <header>
    <id>[unique identifier]</id>
    <topic>$ARGUMENTS</topic>
    <format>[PROBLEM-STATEMENT|REQUIREMENTS|KNOWLEDGE-CORPUS]</format>
  </header>

  <findings>
    <finding confidence="[0.0-1.0]" epistemic="[VERIFIED|STATED|INFERRED|SPECULATIVE]">
      [Key finding or requirement]
    </finding>
    <!-- ... more findings ... -->
  </findings>

  <assumptions>
    <assumption status="[confirmed|challenged|unknown]">
      [Assumption that was surfaced]
    </assumption>
  </assumptions>

  <gaps>
    <gap priority="[high|medium|low]">
      [Information still needed]
    </gap>
  </gaps>

  <next-steps>
    1. [Recommended follow-up action]
    2. [Additional research needed]
  </next-steps>
</interview-output>

Quality Gates

  • Topic is clearly defined and scoped
  • Key assumptions have been surfaced and examined
  • Confidence levels assigned to all findings
  • Information gaps explicitly identified
  • Next steps actionable and specific
  • Output ready for research-brief input

Workflow Integration

This skill is the first step in the research pipeline:

/research-interview → /research-brief → /consolidate-research
     (elicit)            (design)           (synthesize)

After completing a research interview, consider:

  • Run /research-brief with the problem statement to design multi-LLM research
  • Run /evaluate-schema if the topic involves data modeling
  • Run /compare-options if multiple solutions were identified