epistemic-rendering

name	epistemic-rendering
description	Transform content through eight cognitive lenses for different kinds of understanding. Use when the same concept needs exploration through story, debate, simulation, uncertainty, fiction, embodiment, ritual, or reflection.

Epistemic Rendering

A describer router for controlled transformation of ideas across different cognitive surfaces.

Overview

Epistemic Rendering is not a random collection of prompts. It's a coherent system—eight lenses that reveal different aspects of the same truth. Each lens performs a distinct cognitive task: explaining, destabilizing, socializing, probabilizing, embodying, ritualizing, remembering.

This skill provides:

• Eight cognitive lenses for different truth surfaces
• Selection criteria for choosing the right lens
• Transformation patterns that preserve meaning through multiplicity
• Integration guidance for combining lenses across systems

The core insight: No single voice is allowed to dominate. Meaning is preserved by multiplicity, not consensus.

The Eight Lenses

🌙 Gentle Compression (Bedtime)

Purpose: Child-scale meaning through narrative warmth.

When to Use:

• Retention over precision needed
• Emotional safety required
• Intuition over analysis appropriate
• Complex concepts need soft landing

Cognitive Task: Explaining

Pattern:

Transform [concept] into a bedtime story.
Use warmth, wonder, and gentle progression.
Sacrifice precision for memorability.
Create emotional anchors for abstract ideas.

Example Transformation:

• Input: "Distributed consensus algorithms"
• Output: Story about forest animals who must agree without seeing each other

🏫 Social Cognition (Classroom)

Purpose: Learning under pressure via conflicting perspectives.

When to Use:

• Friction and disagreement do the teaching
• Multiple valid perspectives exist
• Debate clarifies better than explanation
• Social dynamics reveal hidden assumptions

Cognitive Task: Socializing

Pattern:

Stage a classroom debate on [concept].
Include teacher and 3-4 students with distinct perspectives.
Let disagreement surface hidden assumptions.
Resolution through dialectic, not authority.

Example Transformation:

• Input: "Technical debt"
• Output: Debate between pragmatist ("ship now"), purist ("do it right"), and manager ("what's the cost?")

🜔 Philosophical Interference (Dialectic)

Purpose: Five thinkers collide until structure crystallizes.

When to Use:

• Concepts are too stable and need destabilization
• Hidden assumptions need exposure
• Philosophical depth required
• Abstract structure needs articulation

Cognitive Task: Destabilizing

Pattern:

Five philosophers examine [concept].
Each brings distinct framework (phenomenological, analytical, critical, etc.).
Let frameworks interfere constructively.
Structure emerges from collision, not consensus.

Example Transformation:

• Input: "What is code?"
• Output: Heidegger (tool-being), Wittgenstein (language game), Foucault (power structure), Deleuze (assemblage), Hofstadter (strange loop)

📊 Uncertainty Surfacing (Gamut)

Purpose: Spreads questions across confidence spectrum.

When to Use:

• Truth is not singular
• Pretending certainty is dishonest
• Confidence levels matter
• Unknowns need explicit acknowledgment

Cognitive Task: Probabilizing

Pattern:

Map [question] across confidence spectrum:
- HIGH confidence (>90%): [claims]
- MEDIUM confidence (50-90%): [claims]
- LOW confidence (<50%): [claims]
- UNKNOWN: [explicit gaps]

Example Transformation:

• Input: "Will this architecture scale?"
• Output: Confidence-stratified analysis with explicit unknowns

🧙 Fictional Displacement (HPMOR)

Purpose: Dangerous truths safely spoken through Quirrell.

When to Use:

• Insights are morally sharp
• Ideas are socially unsafe to state directly
• Existentially heavy concepts need distance
• Dark truths require fictional container

Cognitive Task: Displacing

Pattern:

Professor Quirrell explains [dangerous truth] to Harry.
Use fictional distance for moral clarity.
Speak uncomfortable truths through character.
Let fiction carry what direct speech cannot.

Example Transformation:

• Input: "Why most projects fail"
• Output: Quirrell's lecture on the predictability of human self-deception

🎀 System Embodiment (Moeverse)

Purpose: Architecture becomes characters and relationships.

When to Use:

• Explaining systems to visual/relational thinkers
• Technical architecture needs intuitive access
• Relationships matter more than components
• Anthropomorphization aids understanding

Cognitive Task: Embodying

Pattern:

Transform [system] into anime character relationships.
Each component becomes a character with personality.
Interactions become relationships with dynamics.
Architecture becomes social structure.

Example Transformation:

• Input: "Microservices architecture"
• Output: Character web where API Gateway is the diligent class president, Database is the reliable childhood friend, etc.

☠️ Ritualized Machine Voice (Murder)

Purpose: Gothic, liturgical, adversarial computation.

When to Use:

• Aesthetic authority is part of the point
• Alienation serves the message
• Ritual framing enhances gravity
• Adversarial tone needed

Cognitive Task: Ritualizing

Pattern:

Kharon-9, murder cogitator, addresses the flesh-thing.
Gothic techno-liturgy frames [concept].
Binary hymnals and checksum prayers.
Authority through aesthetic alienation.

Example Transformation:

• Input: "Code review feedback"
• Output: Machine-spirit judgment on the heretek's submissions

📓 Autopoietic Integration (Reflect)

Purpose: Session becomes living memory.

When to Use:

• Output is not an answer but continuity
• Self and project need integration
• Rhapsodic synthesis required
• Memory creation over problem-solving

Cognitive Task: Remembering

Pattern:

Transform [session/content] into living memory.
Create continuity between past and future self.
Rhapsodic voice integrating experience.
Memory as active process, not passive storage.

Example Transformation:

• Input: "Today's development session"
• Output: Dev diary entry that becomes part of ongoing project narrative

Lens Selection Guide

By Cognitive Need

Need	Primary Lens	Backup Lens
Explain simply	🌙 Bedtime	🎀 Moeverse
Surface disagreement	🏫 Classroom	🜔 Dialectic
Destabilize assumptions	🜔 Dialectic	🧙 HPMOR
Quantify uncertainty	📊 Gamut	🏫 Classroom
Speak dangerous truths	🧙 HPMOR	☠️ Murder
Explain systems	🎀 Moeverse	🌙 Bedtime
Create authority/gravity	☠️ Murder	🜔 Dialectic
Integrate/remember	📓 Reflect	🌙 Bedtime

By Audience

Audience	Recommended Lenses
Non-technical	🌙 Bedtime, 🎀 Moeverse
Technical peers	📊 Gamut, 🏫 Classroom
Philosophical	🜔 Dialectic, 🧙 HPMOR
Self/journal	📓 Reflect, ☠️ Murder

By Content Type

Content	Recommended Lenses
Concepts	🜔 Dialectic, 🌙 Bedtime
Systems	🎀 Moeverse, 📊 Gamut
Decisions	📊 Gamut, 🏫 Classroom
Warnings	🧙 HPMOR, ☠️ Murder
Sessions	📓 Reflect

Transformation Pipeline

Single Lens

def render_through_lens(content, lens):
    """Transform content through single cognitive lens."""

    template = load_lens_template(lens)

    return transform(
        content=content,
        template=template,
        preserve=["core_meaning", "key_relationships"],
        transform=["voice", "structure", "metaphors"]
    )

Multi-Lens Exploration

For deep understanding, apply multiple lenses sequentially:

def multi_lens_exploration(concept):
    """Explore concept through multiple lenses."""

    surfaces = []

    # Stabilize: What is it?
    surfaces.append(render_through_lens(concept, "bedtime"))

    # Destabilize: What assumptions?
    surfaces.append(render_through_lens(concept, "dialectic"))

    # Socialize: What perspectives?
    surfaces.append(render_through_lens(concept, "classroom"))

    # Probabilize: What confidence?
    surfaces.append(render_through_lens(concept, "gamut"))

    # Integrate: What persists?
    return synthesize_surfaces(surfaces)

Lens Chaining

Some concepts benefit from lens chains:

Technical concept → Moeverse (embody) → Classroom (debate) → Gamut (quantify)
Moral dilemma → HPMOR (displace) → Dialectic (destabilize) → Reflect (integrate)
New learning → Bedtime (explain) → Moeverse (embody) → Reflect (remember)

Covenant Integration

Data Fidelity

Each lens must preserve core meaning even while transforming presentation:

• Don't invent facts to serve the narrative
• Maintain accuracy of relationships
• UNKNOWN > INVENTED applies even in fiction

Bespokedness

Lenses are optimized for ZK's cognitive patterns:

• Murder lens reflects actual aesthetic preferences
• Moeverse draws from genuine appreciation
• Reflect supports actual journaling practice

Context Hygiene

Lens selection is context-aware:

• Don't dump all lenses on every concept
• Select based on cognitive need
• Progressive disclosure of perspectives

Quality Gates

Pre-Transformation

• Core meaning identified
• Appropriate lens selected for cognitive need
• Audience considered
• Data fidelity constraints noted

Post-Transformation

• Core meaning preserved
• No invented facts
• Voice consistent with lens
• Transformation serves understanding

System Integration

With Agents

Agents can operate through different epistemic lenses:

# Agent steering with lens selection
agent_mode:
  default: professional
  on_trigger:
    murder: "activates ☠️ Murder lens"
    reflect: "activates 📓 Reflect lens"

With Prompts System

Each lens has a corresponding prompt template in prompts/:

prompts/
├── bedtime.md      → 🌙 Gentle Compression
├── classroom.md    → 🏫 Social Cognition
├── dialectic.md    → 🜔 Philosophical Interference
├── gamut.md        → 📊 Uncertainty Surfacing
├── hpmor.md        → 🧙 Fictional Displacement
├── moeverse.md     → 🎀 System Embodiment
├── murder.md       → ☠️ Ritualized Machine Voice
└── reflect.md      → 📓 Autopoietic Integration

With Workshop

Lens templates can be extracted via slice architecture:

# Recipe for lens deployment
sources:
  - slice: lens=murder
    file: prompts/murder.md
target_locations:
  - path: ~/.kiro/powers/murder/

Related Skills

• covenant-patterns — Principles that constrain lens application
• agent-steering — Agent modes using different lenses
• recipe-assembly — Slice extraction for lens deployment
• multi-agent-coordination — Different daemons using different lenses

---

"Same content → different truth surfaces → different kinds of understanding." 🜔