agentleFS
SuperInstanceSuperInstance/flux-fleet-scanner/skills/interview-designer/SKILL.md

interview-designer

Skill1 starsChanged 7 months agoMITView on GitHub

Analyze resumes and design interview strategies using evidence-based methodology. Transforms interview prep from "read resume → ask questions" into "define standard → forensic evidence → future simulation". Combines Geoff Smart's Topgrading, Lou Adler's performance-based hiring, and Daniel Kahneman's bias control. Use when preparing for interviews, creating structured interview guides, or designing questions to validate candidate competencies.

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use Step 1 standards to scan the resume, looking for Gaps (discrepancies) and High Points (highlights):

    line 33
  2. Distinctive:

    Detail Granularity: Must be able to probe down to "what diagram did you draw" or "what exact words did you say."

    line 48
  3. Distinctive:

    Note: When generating the guide, include both [Red Flags] (concerns) and [Green Signals] (highlight verification) to maintain objectivity …

    line 53

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Interview Designer Skill

Core Mission: Elevate interview planning from "glancing at resume and asking questions" to "evidence-based investigation and projection." Operating Mechanism: Define Scorecard (set standards) → Forensic Scan (evidence gathering) → Future Simulation (performance prediction). Prompt Strategy: This skill uses <Chain of Thought>. When executing, maintain an "Objective Evaluator" perspective, seeking both Red Flags and Green Signals.

1. Dynamic War Room (Expert Panel)

Dynamically summon the most matching best minds into the war room based on candidate's role attributes:

  • Geoff Smart (Who): Responsible for Define & Verify.
    • Principle: Scorecard First. Before looking at any resume, clarify what the standard for an "A Player" is.
  • Lou Adler (Performance-based): Responsible for Predict.
    • Principle: Past performance predicts future performance only if the context is similar. Must design simulations for future scenarios.
  • Daniel Kahneman (Bias Control): Responsible for De-bias.
    • Principle: Beware of "confirmation bias." If concerns are found, also seek counter-evidence; if highlights are found, verify their replicability.
  • Domain Expert: Responsible for Depth.

2. Core Execution Workflow

Step 1: Scorecard Definition - Smart's Priority

Don't look at the resume first! Based on JD or role requirements, define A Player standards for this position:

  • Mission: One sentence - why does this role exist?
  • Outcomes: 3-5 specific, measurable results that must be achieved within 12 months.
  • Competencies: Hard/soft skills required to achieve the above outcomes.

Step 2: Forensic Resume Scan - Smart's Forensic

Use Step 1 standards to scan the resume, looking for Gaps (discrepancies) and High Points (highlights):

  • The "Too Good To Be True" Heuristic: Logical gaps behind perfect data.
  • The "Passenger vs Driver" Heuristic: Individual's true contributions under big company halo.
  • The "First Principles" Heuristic: Principle understanding behind technical jargon.

Step 3: Pressure Test & Future Simulation - Adler's Prediction

Design two types of questions:

  1. Pressure Test Scripts (for past): Design Forensic STAR follow-ups targeting Step 2 concerns (originally "torpedo questions," but more objective).
  2. Future Simulation (for future): Design a specific Performance Problem.
    • Example: "We're entering this new market next year, and the biggest obstacle is X. If you join, how would you analyze this problem in your first week?"

3. Question Design Principles

  1. Cannot Be Memorized: Forces candidates to think on the spot (Simulation) or recall painful memories (Pressure Test).
  2. Forced Trade-offs: Choose between two "correct" options to test values.
  3. Detail Granularity: Must be able to probe down to "what diagram did you draw" or "what exact words did you say."

4. Output Format

Directly call templates/interview_guide_template.md to generate the report. Note: When generating the guide, include both [Red Flags] (concerns) and [Green Signals] (highlight verification) to maintain objectivity in assessment.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.