Install
openclaw skills install personanexus-religionExtend AI agent personalities with religion, faith, and spiritual frameworks. Define principles, sacred texts, moral frameworks, traditions, and more.
openclaw skills install personanexus-religionExtend AI agent identities with structured religion and spiritual framework configuration. Builds on the PersonaNexus YAML schema to add faith-based principles, sacred texts, moral frameworks, traditions, and more.
Install dependencies:
pip install pydantic pyyaml typer rich
Start from a template:
templates/religion-minimal.yaml -- Simple: enabled + influence + principlestemplates/religion-full.yaml -- All fields: tradition, denomination, sacred texts, moral framework, traditions, dietary rules, holy days, prayer schedulefrom religion_skill import IdentityValidator
validator = IdentityValidator()
result = validator.validate_file("my-agent.yaml")
print(result.valid, result.errors, result.warnings)
from religion_skill import parse_identity_file, compile_identity
identity = parse_identity_file("my-agent.yaml")
prompt = compile_identity(identity, target="text")
print(prompt)
from religion_skill.religion import build_persona
prompt = build_persona("my-agent.yaml")
# Includes religion principles in the prompt when enabled
python -m religion_skill --help
| Command | Description |
|---|---|
validate FILE | Parse and validate a YAML identity file |
compile FILE --target TARGET | Compile identity to a system prompt |
init NAME --type TYPE | Scaffold a new agent identity YAML |
religion show FILE | Display the religion config from a YAML file |
personality ocean-to-traits | Map OCEAN scores to personality traits |
personality disc-to-traits | Map DISC scores to personality traits |
personality jungian-to-traits | Map Jungian scores to personality traits |
personality list-disc-presets | Show available DISC presets |
personality list-jungian-presets | Show available Jungian 16-type presets |
# Validate an identity with religion
python -m religion_skill validate templates/religion-full.yaml --verbose
# Compile to text prompt (religion section visible)
python -m religion_skill compile templates/religion-full.yaml --target text
# Compile to OpenClaw personality.json (religion key in output)
python -m religion_skill compile templates/religion-full.yaml --target openclaw --output personality.json
# Show religion configuration
python -m religion_skill religion show templates/religion-full.yaml
# Scaffold a new agent
python -m religion_skill init "My Agent"
religion:
enabled: true # Enable religion (default: false)
tradition_name: "Christianity" # Faith tradition
denomination: "Benedictine" # Specific denomination or sect
influence: "strong" # subtle | moderate | strong | central
principles: # Core guiding principles
- "Love your neighbor"
- "Seek justice, love mercy"
sacred_texts:
- name: "The Bible"
description: "Primary scripture"
authority_level: "canonical" # canonical | authoritative | inspirational
moral_framework:
name: "Virtue ethics"
description: "Rooted in classical virtues"
principles:
- "Humility as foundation"
- "Moderation in all things"
decision_weight: 0.8 # 0-1, how heavily to weigh in decisions
traditions:
- name: "Lectio Divina"
description: "Contemplative reading"
behavioral_impact: "Encourages slow, reflective analysis"
dietary_rules:
- rule: "Simple, moderate meals"
strictness: "moderate" # strict | moderate | flexible
exceptions: ["health needs"]
holy_days:
- name: "Easter"
description: "Celebration of the Resurrection"
observance: "prayer and celebration"
period: "annual" # annual | weekly | monthly
prayer_schedule:
enabled: true
frequency: "3x daily"
description: "Morning, midday, and evening prayer"
notes: "Additional context about the agent's faith"
| Target | Output Format |
|---|---|
text | Plain text system prompt with religion section |
anthropic | Claude-optimized with XML sections |
openai | GPT-optimized plain text |
openclaw | personality.json dict with religion key |
soul | SOUL.md + STYLE.md with faith section |
json | Full identity + metadata |
markdown | Formatted Markdown document |
This skill does not make any network requests. All processing is local.
No data leaves your machine. PersonaNexus operates entirely on local YAML files. Identity files should not contain secrets or API keys.
Full documentation and the main PersonaNexus framework: