mismagent-explore

Guides dialogue-driven domain exploration that turns a raw idea into a product brief and context map.

Updated Jun 12, 2026
One-click install
npx skills add https://github.com/lucolucus/mismagent --skill mismagent-explore-lucolucus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mismagent-explore
Source: https://github.com/lucolucus/mismagent/tree/main/codex/skills/mismagent-explore
Command: npx skills add https://github.com/lucolucus/mismagent --skill mismagent-explore-lucolucus

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Starting a feature by jumping straight into code produces misunderstood problems, inconsistent naming, and premature architecture. This Skill structures the exploration phase of the mismAgent methodology so a raw idea is challenged, modeled, and documented before any contract, task, or code exists. ## Core Features & Use Cases - Challenger subagent: dispatches a fresh-context agent that tries to demolish the idea (KILL / RESHAPE / PROCEED verdicts) before any modeling happens. - Analyst subagent: models bounded contexts and fixes the ubiquitous language, writing and amending the project's context-map.md and tactical-model.md. - Project profile bootstrap: creates .mismagent/profile.md from a template capturing output_dir, validation_mode, materials, and capacity, which every downstream skill reads. - Use Case: You have a one-sentence feature idea for a maintenance SaaS. Run the explore movement to dialogue through goals and constraints, have the challenger attack the idea, and receive a product-brief.md plus an amended context-map.md that the model movement can consume. ## Quick Start Ask the agent to explore a new feature idea by describing it in one sentence and letting it run the challenger and analyst subagents before any planning begins.

Frequently Asked Questions about mismagent-explore

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

FAQPage Schema
How do I start a new feature with the mismAgent explore movement?

Invoke the explore command with your idea in one sentence. The skill dialogues with you, dispatches the challenger to attack the idea, then the analyst to model the domain, producing product-brief.md and context-map.md before any planning.

What artifacts does the explore phase produce?

It produces a product-brief.md (problem, user, value, scope), a context-map.md with bounded contexts and ubiquitous language, a tactical-model.md seed file, optional spikes and research notes, and a bootstrapped .mismagent/profile.md on the first feature.

Can explore write code, contracts, or task lists?

No. Explore explicitly forbids contracts, tasks, and code. Its only outputs are analysis documents consumed by the model movement, enforcing the no-premature-coding rule of the methodology.

What happens if the challenger subagent rejects my idea?

A KILL verdict stops the flow and reports back to you; RESHAPE means you redesign the idea together; PROCEED requires closing the MUST_ANSWER_BEFORE_MODELING items before the analyst models the domain.

Does explore work in read-only or plan mode?

Yes. Dialogue and the read-only challenger proceed normally, but the analyst and researcher are not dispatched until writes are allowed. Pending files are listed in the plan and materialized first once writes reopen.

When should I not split work into multiple features during exploration?

A feature is a unit of delivery, not analysis. When asked for one feature per bounded context, probe whether the user wants depth instead — depth lives in the tactical model, research notes, and rich block files, not in more features.