What Next Colleague

Guide multi-turn research conversations to identify and resolve user bottlenecks.

1|Updated Feb 8, 2026
One-click install
npx skills add https://github.com/soheunyi/get-research-done --skill what-next-colleague
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: What Next Colleague
Source: https://github.com/soheunyi/get-research-done/tree/main/skills/grd-what-next-colleague
Command: npx skills add https://github.com/soheunyi/get-research-done --skill what-next-colleague

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Run a structured, multi-turn conversation to help users decide the best next research direction and concrete next actions.

Core Features & Use Cases

  • Guided questioning loop: asks high-leverage questions, offers concrete options, and records progress.
  • Deterministic handoffs to specialized roles when escalation is needed (e.g., Hypothesis Designer, Experiment Planner).
  • Context-capturing summaries after each turn to maintain alignment and trace decisions.

Quick Start

Begin a guided questioning session by asking one high-leverage question, offering 2-4 concrete options, and summarizing progress until a clear next action is defined.

Frequently Asked Questions about What Next Colleague

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I decide my next research direction when I feel stuck?

To decide your next research direction when stuck, engage in a structured, multi-turn conversation that asks high-leverage questions and offers concrete options to resolve your bottleneck. The guided loop captures progress with periodic summaries until a clear next action is defined.

What is iterative questioning for selecting a next action in research?

Iterative questioning for selecting a next action is a guided loop protocol that asks one high-leverage question per turn and provides 2-4 concrete options. It uses back-and-forth reasoning to resolve objectives, constraints, and success criteria for your research.

How do I guide a structured conversation to plan my next experiment?

You guide a structured conversation to plan experiments by enforcing a questioning loop that clarifies your objective and environment. The process provides an explicit open-ended path and periodic context summaries until success criteria are resolved for a deterministic handoff.

Can I use guided questioning to hand off a hypothesis to an experiment planner?

Yes, you can use guided questioning to hand off a hypothesis to an experiment planner. The conversation resolves your objective, constraints, and success criteria, then performs deterministic handoffs to specialized roles like Hypothesis Designer or Experiment Planner.

What's the best way to resolve research bottlenecks before planning experiments?

The best way to resolve research bottlenecks before planning experiments is through a structured conversation that narrows down your next action. It enforces a protocol of offering concrete options and an open-ended path until your objective, environment, and constraints are fully resolved.

Does this guided conversation approach work for multi-turn research planning?

Yes, this guided conversation approach works for multi-turn research planning by maintaining alignment through context-capturing summaries after each turn. It traces decisions and records progress until the objective, constraints, environment, and success criteria are completely resolved.