guild-loop-clarify

Enforce an adversarial architect-researcher loop to surface assumptions and unresolved questions.

6|Updated Apr 24, 2026

Search Tags:#assumptions #risk-mitigation #guild #clarify #spec-workflow #adversarial-loop #architect-researcher

Ready to run

npx skills add https://github.com/lookatitude/guild --skill guild-loop-clarify

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: guild-loop-clarify
Source: https://github.com/lookatitude/guild/tree/main/skills/meta/loop-clarify
Command: npx skills add https://github.com/lookatitude/guild --skill guild-loop-clarify

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps teams surface missing assumptions and unresolved questions before committing a spec by enforcing an adversarial loop between architect and researcher. It reduces the risk of scope creep and rework by surfacing concerns early.

Core Features & Use Cases

Structured, sentinel-terminated dialogues between architect (writer) and researcher (challenger) that expose gaps before writing the spec.
Enforced workflow controls (cap, sentinel checks, escalation paths) that produce a provable trail of rounds, handoffs, and assumptions.
Use Case: when preparing a guild spec or feature scope, run the loop to gather questions and convert them into explicit assumptions appended to the spec.

Quick Start

Start the adversarial clarify loop before drafting the spec by invoking the guild workflow with the appropriate --loops mode.

Frequently Asked Questions about guild-loop-clarify

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

FAQPage Schema

How do I surface missing assumptions before committing a feature spec?▼

To surface missing assumptions before committing a spec, run an adversarial clarify loop between an architect and researcher role. This structured dialogue enforces a sentinel-terminated process that exposes unresolved questions and gaps early.

What is the best way to mitigate scope creep risk when preparing guild specs?▼

Mitigating scope creep risk for guild specs requires enforcing an adversarial loop with a fixed iteration cap and sentinel termination. This process converts unresolved questions into a consolidated assumptions list appended directly to the spec.

How does an adversarial loop between architect and researcher work for spec clarification?▼

The adversarial clarify loop works by having the architect draft scoping details while the researcher challenges them, creating structured handoffs. It enforces workflow controls, tracks rounds, and escalates unresolved questions before spec writing begins.

Can I enforce a fixed iteration limit when clarifying cross-team alignment questions?▼

Yes, you can enforce a fixed iteration limit for cross-team alignment clarification. The loop applies a strict cap with sentinel checks to terminate the dialogue, producing a provable trail of rounds and handoffs for the spec.

When do I need to run a spec clarification loop for cross-team alignment?▼

You need to run a spec clarification loop when preparing guild specs that require cross-team alignment across iterations. It is specifically designed to gather unresolved questions and mitigate risks before you commit to writing the final spec.

What outputs should I expect from running an adversarial spec clarification workflow?▼

Running an adversarial spec clarification workflow produces outputs including status, completed rounds, handoffs, unresolved questions, and a consolidated assumptions list for the spec. This provides a provable trail for risk mitigation.