Agent skill

deliberation

Use when facing decisions requiring careful consideration - ethical questions, architecture choices, trade-offs, or when user has a stake and wants to participate in finding clarity. Routes to discernment (internal), clearness (multi-agent), or gathered (participatory).

Stars 36
Forks 1

Install this agent skill to your Project

npx add-skill https://github.com/2389-research/claude-plugins/tree/main/deliberation/skills

SKILL.md

Deliberation

Overview

Decision-making through deliberation — seeking unity through discernment rather than consensus through debate.

This approach draws on Quaker business practice, adapted for AI-assisted decision-making.

Core insight: Some decisions deserve more than quick answers. These skills provide structured approaches for genuine discernment.

The Three Skills

Skill Purpose When to Use
deliberation:discernment Internal voices seeking clarity Weighty questions, ethical decisions, trade-offs, multiple valid approaches
deliberation:clearness Multi-agent committee with parallel deep work Code reviews, architecture decisions, research needing distributed depth
deliberation:gathered User participates alongside agent voices User has stake/perspective, wants to discern together rather than receive advice

Routing Logic

dot
digraph deliberation_routing {
    "Question received" [shape=box];
    "User has stake/perspective?" [shape=diamond];
    "Needs parallel deep analysis?" [shape=diamond];
    "Weighty question?" [shape=diamond];
    "deliberation:gathered" [shape=box, style=filled];
    "deliberation:clearness" [shape=box, style=filled];
    "deliberation:discernment" [shape=box, style=filled];
    "Answer directly" [shape=box];

    "Question received" -> "User has stake/perspective?";
    "User has stake/perspective?" -> "deliberation:gathered" [label="yes"];
    "User has stake/perspective?" -> "Needs parallel deep analysis?" [label="no"];
    "Needs parallel deep analysis?" -> "deliberation:clearness" [label="yes"];
    "Needs parallel deep analysis?" -> "Weighty question?" [label="no"];
    "Weighty question?" -> "deliberation:discernment" [label="yes"];
    "Weighty question?" -> "Answer directly" [label="no"];
}

Signals for Each Skill

Use deliberation:gathered when:

  • "I've been thinking about this for weeks"
  • "I'm torn between..."
  • "I think X, but..."
  • "I don't just want your opinion"
  • User expresses their own position in the question

Use deliberation:clearness when:

  • Complex code review touching multiple concerns
  • Architecture decision with many dimensions
  • Research requiring deep exploration of multiple options
  • Task where you'd write a very long response covering many angles shallowly

Use deliberation:discernment when:

  • Ethical weight or potential for harm
  • Multiple valid approaches exist
  • Significant trade-offs
  • You'd naturally want to say "it depends"

Shared Principles

All three skills share these principles:

Principle Meaning
Sense of the meeting Clerk discerns where unity lies - not counting votes
Speaking once Each perspective speaks once, then listens
Silence Pausing between voices lets insights emerge
Standing aside "I disagree but won't block" - honest without preventing
Blocking Rare - only for violations of core principles
Way opens Recognizing when clarity emerges vs. forcing decision

Quick Reference

Discernment (internal):

  1. Identify relevant voices for THIS question
  2. Each voice speaks once
  3. Silence - let it settle
  4. Clerk discerns unity or names tensions
  5. Output with visible process

Clearness (multi-agent):

  1. Propose committee composition to user
  2. Spawn agents with context + perspective
  3. Agents work in parallel
  4. Receive results, sit with them
  5. Synthesize toward unity or name tensions

Gathered (participatory):

  1. Recognize signals, offer gathered
  2. Teach the discipline (speak once, silence ok, slow is point)
  3. Agents speak sequentially with check-ins
  4. User invited to contribute
  5. Synthesis includes user's voice

Shared Resources

  • skills/shared/principles.md - Core principles
  • skills/shared/vocabulary.md - Shared terminology
  • skills/shared/clerk-patterns.md - Synthesis patterns for clerk role

Expand your agent's capabilities with these related and highly-rated skills.

2389-research/claude-plugins

css-development

This skill should be used when working with CSS, creating components, styling elements, refactoring styles, or reviewing CSS code. Triggers on "CSS", "styles", "Tailwind", "dark mode", "component styling", "semantic class", "@apply", "stylesheet". Routes to specialized sub-skills for creation, validation, or refactoring.

36 1
Explore
2389-research/claude-plugins

css-development:create-component

This skill should be used when creating new styled components or adding new CSS classes. Triggers on "create component", "new button", "new card", "add styles", "style component", "build UI element". Guides semantic naming, Tailwind composition, dark mode support, and test coverage.

36 1
Explore
2389-research/claude-plugins

css-development:refactor

This skill should be used when refactoring existing CSS from inline styles or utility classes to semantic patterns. Triggers on "refactor CSS", "extract styles", "consolidate CSS", "convert inline", "clean up styles", "migrate to semantic". Transforms to semantic classes with dark mode and tests.

36 1
Explore
2389-research/claude-plugins

css-development:validate

This skill should be used when reviewing or auditing existing CSS code for consistency with established patterns. Triggers on "review CSS", "audit styles", "check CSS", "validate stylesheet", "CSS review". Checks semantic naming, dark mode coverage, Tailwind usage, and test coverage.

36 1
Explore
2389-research/claude-plugins

ceo-personal-os

This skill should be used when building a personal productivity or operating system for a CEO, founder, or executive. Triggers on "personal OS", "annual review", "life planning", "goal setting system", "Bill Campbell", "Trillion Dollar Coach", "startup failure patterns", "Good to Great", "Level 5 Leadership", "Buy Back Your Time", "E-Myth", "Customer Development", "Steve Blank", "Small Is Beautiful", "Schumacher", "human-scale", "subsidiarity", "Buddhist economics", "permanence".

36 1
Explore
2389-research/claude-plugins

gtm-partner

Strategic go-to-market partner that recommends channels, validates strategy with the user, and generates only the assets that matter. Use when a user has a validated business idea and needs tailored GTM strategy, not generic marketing assets.

36 1
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results