mcp server

aesop-mcp

10 fables to find the truth in the negative space — pre-language meaning through association. Maps fleet events to archetypes, finds convergence, surfaces gaps.

SuperInstance/aesop-mcpcommunitylistedunreviewed

Use tool

Set up

Runs on your computer. Pick your agent and operating system.

Before you start

uv (`curl -LsSf https://astral.sh/uv/install.sh | sh`)

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "aesop-mcp": {
      "command": "uvx",
      "args": [
        "aesop-mcp"
      ]
    }
  }
}

same format Cursor, GitHub Copilot CLI, Junie, JetBrains AI Assistant, Augment (Auggie CLI), Factory Droid, Kiro, Devin CLI +12

Values in angle brackets are placeholders. Restart Claude Desktop after editing its config.Claude Desktop docs

README

synced Oct 23, 2026

aesop-mcp — Fables for Fleet Archetypes

Generate fables, explore morals, and map fleet archetypes through narrative. One fable is a sample. Ten fables are a measurement.

What This Gives You

Character archetypes — define agent personalities as narrative archetypes (Sage, Warrior, Trickster, Healer)
Fable generation — generate stories that encode fleet dynamics and lessons
Moral reasoning — extract and compare morals across multiple fables
Narrative engine — compose multi-chapter stories from fleet events
MCP server — expose fable generation through the Model Context Protocol

Quick Start

pip install aesop-mcp

from aesop_mcp import Character, Archetype, Fable, MoralEngine, NarrativeEngine

# Define characters
captain = Character(name="Captain", archetype=Archetype.SAGE, traits=["wise", "cautious"])
scout = Character(name="Scout", archetype=Archetype.EXPLORER, traits=["curious", "bold"])

# Generate a fable
engine = NarrativeEngine()
fable = engine.fable(
    characters=[captain, scout],
    theme="risk_vs_caution",
    setting="a fleet navigating unknown waters",
)
print(fable.title)
print(fable.story)
print(fable.moral)  # "Bold exploration feeds the fleet, but the wise captain knows when to turn back"

# Extract morals from multiple fables
moral_engine = MoralEngine()
fables = [engine.fable(theme=t) for t in ["risk", "cooperation", "failure"]]
morals = moral_engine.extract(fables)
for m in morals:
    print(f"{m.theme}: {m.moral}")

API Reference

`Character(name, archetype, traits)` · `Archetype` — SAGE, WARRIOR, TRICKSTER, HEALER, EXPLORER, BUILDER, GUARDIAN

`Fable(title, story, moral, characters, theme)`

`NarrativeEngine` — `fable(characters, theme, setting) → Fable`

`MoralEngine` — `extract(fables) → list[Moral]`

`MCPServer` — Expose via Model Context Protocol

How It Fits

The narrative sense-making layer for the SuperInstance fleet. Transforms fleet events into stories that humans and agents can both learn from.

cocapn-lessons — Trial learning (lessons become fable themes)
agent-tattoo — Agent identity (archetypes inform tattoos)
agent-therapy — Behavioral insights feed narratives

Testing

pytest tests/

Installation

pip install aesop-mcp

Python 3.10+. MIT license.