← 返回
AI智能 中文

Personal Board of Directors

Add a Historical Figures Advisory Board to AI agent personalities. 10 pre-configured personas inspired by public-domain historical figures for strategic advice.
为AI代理角色添加历史人物顾问委员会,提供10个基于公共领域历史人物的预置人格,用于提供战略建议。
jcrowan3
AI智能 clawhub v1.0.0 1 版本 100000 Key: 无需
★ 1
Stars
📥 484
下载
💾 18
安装
1
版本
#latest

概述

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:

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

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

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

from board_skill.board import build_persona

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

CLI

python -m board_skill --help

CLI Commands

CommandDescription
----------------------
validate FILEParse and validate a YAML identity file
compile FILE --target TARGETCompile identity to a system prompt
init NAME --type TYPEScaffold a new agent identity YAML
board show FILEDisplay the board config from a YAML file
board list-members FILEList all board members with personality details
personality ocean-to-traitsMap OCEAN scores to personality traits
personality disc-to-traitsMap DISC scores to personality traits
personality jungian-to-traitsMap Jungian scores to personality traits
personality list-disc-presetsShow available DISC presets
personality list-jungian-presetsShow available Jungian 16-type presets

CLI Examples

# 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

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

TargetOutput Format
-----------------------
textPlain text system prompt with board section
anthropicClaude-optimized with XML sections
openaiGPT-optimized plain text
openclawpersonality.json dict with board key
soulSOUL.md + STYLE.md with advisory section
jsonFull identity + metadata
markdownFormatted 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

版本历史

共 1 个版本

  • v1.0.0 当前
    2026-03-30 10:22 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-intelligence

self-improving agent

pskoett
捕获经验教训、错误和纠正,以实现持续改进。使用时机:(1)命令或操作意外失败;(2)用户纠正……
★ 4,055 📥 796,000
ai-intelligence

ontology

oswalpalash
类型化知识图谱,用于结构化智能体记忆与可组合技能。支持创建/查询实体(人员、项目、任务、事件、文档)及关联...
★ 709 📥 243,531
ai-intelligence

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,349 📥 317,701