Personal Board of Directors

jcrowan3@jcrowan3

Add a Historical Figures Advisory Board to AI agent personalities. 10 pre-configured personas inspired by public-domain historical figures for strategic advice.

Install

openclaw skills install @jcrowan3/personanexus-bod

PersonaNexus Board of Directors Skill

Extend AI agent identities with a structured Board of Directors advisory panel composed of historical figures. Builds on the PersonaNexus YAML schema to add board members, engagement rules, and strategic advisory capabilities.

What This Skill Does

Define agent advisory board configuration via YAML
Compile board context into system prompts alongside personality traits
Validate board configuration with semantic warnings
Export to OpenClaw personality.json, SOUL.md, Anthropic XML, and more
Pre-configured 10 historical figures ready to use

Setup

Install dependencies:

bash

pip install pydantic pyyaml typer rich

Usage

1. Create an Agent with a Board

Start from a template:

templates/board-minimal.yaml -- Simple: enabled + 2 members + basic engagement
templates/board-full.yaml -- All fields: 10 historical figures, full engagement rules, disclaimer

2. Validate

python

from board_skill import IdentityValidator

validator = IdentityValidator()
result = validator.validate_file("my-agent.yaml")
print(result.valid, result.errors, result.warnings)

3. Compile to a System Prompt

python

from board_skill import parse_identity_file, compile_identity

identity = parse_identity_file("my-agent.yaml")
prompt = compile_identity(identity, target="text")
print(prompt)

4. Quick build_persona Helper

python

from board_skill.board import build_persona

prompt = build_persona("my-agent.yaml")
# Includes board members in the prompt when enabled

CLI

bash

python -m board_skill --help

CLI Commands

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
`board show FILE`	Display the board config from a YAML file
`board list-members FILE`	List all board members with personality details
`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

CLI Examples

bash

# Validate an identity with board
python -m board_skill validate templates/board-full.yaml --verbose

# Compile to text prompt (board section visible)
python -m board_skill compile templates/board-full.yaml --target text

# Compile to OpenClaw personality.json (board key in output)
python -m board_skill compile templates/board-full.yaml --target openclaw --output personality.json

# Show board configuration
python -m board_skill board show templates/board-full.yaml

# List board members with details
python -m board_skill board list-members templates/board-full.yaml

# Scaffold a new agent
python -m board_skill init "My Agent"

Board YAML Schema

yaml

board:
  enabled: true                          # Enable board (default: false)
  disclaimer: "These are fictional..."   # Required disclaimer text

  board_members:
    - id: "sun_tzu"
      name: "Sun Tzu"
      historical_figure: "Sun Tzu"
      died: "~496 BC"
      board_role: "Chief Strategist"
      core_mindset: "Win without fighting"
      modern_relevance: "Competitive strategy"
      personality:
        ocean:
          openness: 0.9
          conscientiousness: 0.95
          extraversion: 0.4
          agreeableness: 0.3
          neuroticism: 0.2
        disc_style: "dominance"          # dominance | influence | steadiness | compliance
        jungian_type: "INTJ"
      key_quote: "The supreme art of war is to subdue the enemy without fighting"

  engagement_rules:
    - "Each member responds from their historical lens"
    - "Synthesize consensus or majority vote"

Compile Targets

Target	Output Format
`text`	Plain text system prompt with board section
`anthropic`	Claude-optimized with XML sections
`openai`	GPT-optimized plain text
`openclaw`	personality.json dict with board key
`soul`	SOUL.md + STYLE.md with advisory section
`json`	Full identity + metadata
`markdown`	Formatted Markdown document

External Endpoints

This skill does not make any network requests. All processing is local.

Security & Privacy

No data leaves your machine. PersonaNexus operates entirely on local YAML files. Identity files should not contain secrets or API keys.

Learn More

Full documentation and the main PersonaNexus framework:

https://github.com/PersonaNexus/personanexus