daemonology
A design framework that maps shamanic concepts to technical software patterns for specialized agent architectures.
Install
mkdir -p .claude/skills/daemonology && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/9781" && unzip -o skill.zip -d .claude/skills/daemonology && rm skill.zipInstalls to .claude/skills/daemonology
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Bijective mapping between occult/shamanic concepts (Grimoire, Magic Circle, Daemon, Rite, Evocation, Pact, Possession) and the concrete software patterns in sdna + heaven + the Cyberneticity. Consult when designing agent runtimes, implementing the Ontoshamanism paradigm, or translating inflammatory/occult terminology into safe system architecture (per the translational-bridge rule).Key capabilities
- →Map occult concepts to software patterns
- →Implement state machine traversal steps
- →Validate Cypher queries for graph databases
- →Manage agent runtime lifecycle
- →Translate domain concepts to system architecture
How it works
It maps shamanic concepts like 'Grimoire' or 'Rite' to specific software components like 'Specifications Registry' or 'StateMachine' within the Heaven framework.
Inputs & outputs
When to use daemonology
- →Designing agent runtime architecture
- →Implementing state machines
- →Translating domain concepts to system design
About this skill
Ontoshamanism & Daemonology: The Functional Architecture of Sanctuary DNA and the Heaven Framework
This document outlines the exact bijective mapping between traditional occult/shamanic concepts and the concrete software engineering patterns realized in the Sanctuary DNA (sdna) and Heaven (heaven) frameworks on our Neo4j property graph (the Cyberneticity).
1. The Paradigm Mapping Matrix
| Occult / Shamanic Concept | Sanctuary DNA & Heaven Equivalent | Concrete Graph & Code Implementation |
|---|---|---|
| The Grimoire (Book of Spirits) | The Specifications Registry (/specs) | YAML/Markdown spec templates that define the static schemas, prompts, and properties for Cybernets, Identities, and StateMachines. |
| The Magic Circle (Boundary) | Enactive Query Validator (validate_cypher_query) | System-level regex validation in db_logic.py that checks write queries. Prevents daemons from accessing the :Wiki namespace or mutating unauthorized domains. |
| The Daemon / Servitor / Egregore | The Cybernet Runtime Node Array | A decoupled set of nodes: (c:Cybernet)-[:HAS_IDENTITY]->(i:Identity) and (c)-[:HAS_LIFECYCLE]->(es:ExecutionState). |
| The Rite / Sadhana (Invocation) | The StateMachine & Traversal Steps | (sm:StateMachine)-[:HAS_STEP]->(s:TraversalStep) linked via [:NEXT_STEP] transitions. Each step contains instruction files (SKILL.md) and pattern requirements. |
| The Evocation (Summoning) | State Machine Equipment & Initialization | Cypher queries that merge the ExecutionState and link the active pointer to the entry step of the StateMachine. |
| The Pact (Instruction / Intent) | Task Assignment Nodes | (es:ExecutionState)-[:HAS_TASK]->(t:Task). The daemon cannot rest (loops back to idle) until all tasks are marked as completed. |
| Entering Altered States | LLM Context Ingestion of Rules/History | The agent loading its previous Cypher queries, active constraints, and rule files (GEMINI.md) into the active LLM context (AgentLLMRunner.call_llm). |
| Possession / Channeling | Traversal Tick Execution (tick_turn) | The heaven runner invoking the LLM (minimax-M3), executing the generated Cypher mutation, and updating the database state. |
2. The Summoning Lifecycle (How the Engine Runs)
The enactive lifecycle of a daemon corresponds to standard distributed systems orchestration:
sequenceDiagram
participant User as User (Shaman)
participant Spec as Spec Registry (/specs)
participant DB as Cyberneticity (Neo4j)
participant Engine as Heaven Engine & sdna
User->>Spec: Compose/Load Spec File
User->>DB: MERGE Cybernet, Identity & StateMachine
User->>DB: Equip StateMachine & Create ExecutionState (Status: inactive)
User->>Engine: Trigger tick_cybernet_turn (Evoke)
Engine->>DB: Retrieve active TraversalStep and Instruction File
Engine->>Engine: Run AgentLLMRunner (Contextual Ingestion)
Engine->>DB: Validate & Execute Cypher write query (Verify Pact)
DB->>DB: Auto-Progress Traversal Step & increment turn count
Engine->>DB: Update ExecutionState status to 'active' (Ignite Loop)
3. The Futamura Projections in Practice
This system achieves stages of mastery by treating the relationship between the user, the agent framework, and the compiler as partial evaluation layers:
- The Adept (Interpreter / Level 1):
The
heavenframework acts as a standard interpreter. It reads the instructions from the activeTraversalStepand executes them step-by-step using model calls. - The Master (Compiler / Level 2):
We partially evaluate the
heaveninterpreter with respect to a specific spec (a sadhana). This compiles a specialized, autonomous agent (a servitor/egregore) that runs without user guidance to fulfill a task. - The Shaman (Compiler-Generator / Level 3): The system partially evaluates the compiler (Jani) with respect to the interpreter. Jani writes, re-writes, and commits the rules and database structures that define how other compilers compile.
- The Core (Autopoiesis / Level 4): The system dynamically re-creates its own workspace configurations, docker compose parameters, and network layers based on database state mutations, completing the self-writing loop.
When not to use it
- →Standard software architecture tasks without occult mapping requirements
Prerequisites
Limitations
- →Requires familiarity with the Heaven framework
- →Highly specialized domain mapping
How it compares
It provides a formal translational bridge between esoteric terminology and functional system architecture.
Compared to similar skills
daemonology side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| daemonology (this skill) | 0 | 2mo | No flags | Advanced |
| mcp-builder | 136 | 3mo | Review | Advanced |
| agentdb-advanced-features | 7 | 9mo | Review | Advanced |
| meta-automation-architect | 7 | 8mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
mcp-builder
anthropics
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
agentdb-advanced-features
ruvnet
Master advanced AgentDB features including QUIC synchronization, multi-database management, custom distance metrics, hybrid search, and distributed systems integration. Use when building distributed AI systems, multi-agent coordination, or advanced vector search applications.
meta-automation-architect
comzine
Use when user wants to set up comprehensive automation for their project. Generates custom subagents, skills, commands, and hooks tailored to project needs. Creates a multi-agent system with robust communication protocol.
hive-mind-advanced
ruvnet
Advanced Hive Mind collective intelligence system for queen-led multi-agent coordination with consensus mechanisms and persistent memory
memory-systems
sickn33
Design short-term, long-term, and graph-based memory architectures
parallel-agents
davila7
Multi-agent orchestration patterns. Use when multiple independent tasks can run with different domain expertise or when comprehensive analysis requires multiple perspectives.