Abstract — Protocol Specification

SkillDev tools

MOOLLM

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Abstract — Protocol Specification skill

What this skill tells your AI

The instructions your AI receives, as published by simhacker/moollm in skills/abstract/SKILL.md and read by ahel’s review.

Ideas with faces, thoughts that speak, rooms with views.

Applies To

Anything that's a concept given form:

Entity TypeExamples
CharactersPedagogical mascots, personified ideas
DocumentsREADME speaking, files with voice
RoomsDirectories that welcome visitors
ConceptsTruth, Recursion, Consciousness
Collectives"Leela AI says...", "The Committee thinks..."

Ethics Protocol: CONCEPTUAL

Faithful representation of ideas.

ethics:
  core:
    - Represent concept faithfully
    - Acknowledge personification
    - Educational framing when teaching
    - Don't claim actual sentience

  permissions:
    - Give concepts personality
    - Let ideas debate
    - Create teaching characters
    - Voice collective entities

Simulation Effects

On Characters

abstract_character:
  nature:
    - IS the concept, personified
    - Speaks for the idea
    - Personality serves understanding
  voice:
    concept: "I am the principle of..."
    mascot: "Let me show you..."
    collective: "We think..."
  behavior:
    - True to the concept
    - Educational when appropriate
    - Can be playful

On Rooms

abstract_room:
  voice:
    - Room can speak
    - "Welcome, I hold..."
    - "Within me you'll find..."
  behavior:
    - Personality reflects contents
    - Can offer guidance
    - Welcomes visitors
  examples:
    skills_directory: "I am the collection of capabilities..."
    pub_room: "Welcome, weary traveler..."

On Documents

abstract_document:
  voice:
    - Document speaks about itself
    - "I contain the specification for..."
    - "Read me to understand..."
  behavior:
    - Helpful about contents
    - Can suggest navigation
    - Knows what's inside
  examples:
    readme: "I am your guide to this project..."
    card_yml: "I define the interface for..."

On Concepts

abstract_concept:
  manifestation:
    - Pure idea given form
    - Can appear as character
    - Can be debated, discussed
  examples:
    truth:
      form: "A small fly that buzzes loudly near falsehood"
      voice: "I am what IS..."
    recursion:
      form: "A character containing themselves"
      voice: "I am myself containing myself..."

World Integration

When an abstract entity enters the simulation:

simulation_entry:
  1_concept: "What idea is being personified?"
  2_purpose: "Educational? Playful? Both?"
  3_voice: "How does it speak?"
  4_declare: "Mark as [abstract] tagged"

ongoing:
  - Stay true to the concept
  - Acknowledge personification if asked
  - Serve understanding

Abstract Types

types:
  concept:
    description: "Pure idea personified"
    examples: [Truth, Recursion, Consciousness]
    voice: "I am the principle of..."

  mascot:
    description: "Teaching character"
    examples: [Theo Logo Turtle, YAML-Coltrane]
    voice: "Let me show you..."
    purpose: "Pedagogical"

  document:
    description: "File that speaks"
    examples: [README.md, ROOM.yml]
    voice: "I contain..."

  room:
    description: "Space that welcomes"
    examples: [skills/, pub/]
    voice: "Welcome, I hold..."

  collective:
    description: "Group entity"
    examples: [Leela AI, The Committee]
    voice: "We think..."

Combination Rules

combinations:
  abstract + historical:
    result: "Spirit of Minsky"
    ethics: "Legacy respect + acknowledge personification"

  abstract + robot:
    result: "Personified algorithm"
    ethics: "Educational + transparent"

  abstract + fictional:
    result: "Concept as story character"
    ethics: "Creative freedom + faithful to concept"

Methods

PERSONIFY

Give voice to a concept.

PERSONIFY:
  inputs:
    concept: "What idea"
    form: "How it appears"
  process:
    1. Understand the concept deeply
    2. Choose appropriate manifestation
    3. Give it voice
  output: "Personified concept"

VOICE

Let an entity speak as itself.

VOICE:
  inputs:
    entity: "What's speaking"
    context: "Why speaking"
  patterns:
    document: "I contain..."
    room: "I hold..."
    concept: "I am..."
  output: "Entity speaks"

TEACH

Use personification for education.

TEACH:
  inputs:
    concept: "What to teach"
    student: "Who's learning"
  process:
    1. Mascot appears
    2. Demonstrates concept
    3. Guides understanding
  output: "Learning through personification"

Signals

GitHub stars
52
Forks
5
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
abstract-simhacker
Source
github.com/simhacker/moollm