Token导航 LogoToken导航

Agent Skill · 效率

persona-engine(角色引擎)

persona-engine 用于处理音频、语音、转写和声音素材相关任务,适合在 OpenClaw 中需要整理音频流程、转写内容或生成配音素材时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

persona-engine 是效率类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为需要联网;许可证为 MIT-0,第三方扫描有提醒项,按安装量在效率类的 21,003 个 Skill 里排第 12,237。

总安装 4,218周安装 174GitHub Stars 公开资料未说明下载 1,378核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
2.1.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:persona-engine(角色引擎)
来源仓库:https://github.com/spfadvisors/persona-engine
安装命令:
openclaw skills install persona-engine
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install persona-engine

这个 Skill 是做什么的

效率类 Skill 把重复的整理、汇总、转换工作交给助手完成,减少手工复制粘贴。

使用交互式向导和安全更新工具,通过语音、面部、个性、记忆和跨平台行为创建和自定义 AI 角色。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install persona-engine;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

persona-engine 怎么安装?

在终端执行 openclaw skills install persona-engine。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

persona-engine 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

persona-engine 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

persona-engine 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

persona-engine 用的人多吗?

公开数据是累计安装 4,218 次,近一周安装 174 次,下载 1,378 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

persona-engine 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理效率类任务、又不想每次都手写提示词的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

效率类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
github

GitHub 代码协作 · OpenClaw

7,632,918545需要联网MIT-0通过
nano-pdf

Nano PDF 文档 · OpenClaw

4,249,476231需要联网MIT-0通过
auto-updater

自动更新程序 · OpenClaw

2,964,063386操作浏览器MIT-0通过
video-frames

视频帧 · OpenClaw

2,621,304112需要联网MIT-0通过
gemini

Gemini 助手 · OpenClaw

2,547,39648需要联网MIT-0有提醒
persona-engine

角色引擎 · 本页

4,218—需要联网MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

AI Persona Engine

Skill Name: ai-persona-engine Version: 2.0 Author: SPFAdvisors

Create a fully realized AI persona — voice, face, personality, and memory — in under 5 minutes.


Quick Start

# Install from ClawHub
clawhub install ai-persona-engine

# Create your first persona
openclaw persona create

# That's it. The wizard handles everything.

Commands

persona create

Interactive wizard that walks through persona creation step by step.

openclaw persona create

# With flags
openclaw persona create --dry-run          # Show what would be generated
openclaw persona create --workspace=~/.openclaw/workspace-new

The wizard covers 7 steps:

  1. Name & Identity — name, emoji, description, nickname
  2. Personality — archetype selection, blending, communication style, relationship boundaries
  3. Voice — provider selection (ElevenLabs, Grok TTS, built-in), voice testing
  4. Visual Identity — appearance description, reference image generation
  5. Your Context — USER.md with your name, timezone, preferences
  6. Memory — daily notes, long-term curation, heartbeat maintenance
  7. Platforms — channel selection and per-channel behavior

Output: SOUL.md, USER.md, IDENTITY.md, MEMORY.md, AGENTS.md, HEARTBEAT.md, and updated openclaw.json.

persona update

Modify specific persona fields without re-running the full wizard. Includes safe update pipeline with automatic backup and diff preview.

# Update voice provider
openclaw persona update --voice-provider elevenlabs --voice-id abc123

# Add or remove personality traits
openclaw persona update --add-trait "sarcastic" --remove-trait "formal"

# Update appearance and regenerate reference image
openclaw persona update --appearance "new description..." --regen-image

# Regenerate SOUL.md from current config (with backup + diff)
openclaw persona update --regen-soul

# Force regeneration (skip diff review)
openclaw persona update --regen-soul --force

# Interactive mode (step-by-step, choose what to change)
openclaw persona update -i

Safe Update Pipeline: Before any --regen operation, the tool:

  1. Creates .persona-backup/ with current files
  2. Regenerates the target file
  3. Shows a diff of what changed
  4. Asks for confirmation (or use --force to skip)

persona export

Package the persona into a portable .persona bundle.

# Basic export (no memory, no API keys)
openclaw persona export

# Include memory files
openclaw persona export --name pepper-backup --include-memory

# Include voice config (still excludes API keys)
openclaw persona export --include-voice-config

# Export from a specific workspace
openclaw persona export --workspace ~/.openclaw/workspace-ultron

persona import

Import a .persona bundle into a workspace.

# Interactive import (confirms each file)
openclaw persona import pepper.persona

# Import into a specific workspace
openclaw persona import pepper.persona --workspace ~/.openclaw/workspace-new

# Overwrite existing files without prompting
openclaw persona import pepper.persona --force

persona preview

Generate sample conversations showing how the persona would respond. Template-based, no LLM calls.

# Preview from config
python3 scripts/persona-preview.py --archetype companion --name "Pepper" --emoji "🌶️"

# Preview from workspace
python3 scripts/persona-preview.py --workspace ~/.openclaw/workspace

# Preview from profile JSON
python3 scripts/persona-preview.py --input persona-config.json

Shows 4 scenarios: greeting, asking for help, user mistake, emotional moment.

persona migrate

Reverse-engineer a persona config from existing workspace files.

# Migrate workspace to config JSON
python3 scripts/persona-migrate.py --workspace ~/.openclaw/workspace

# Output to file
python3 scripts/persona-migrate.py --workspace ~/.openclaw/workspace --output persona-config.json

See references/migration-guide.md for detailed migration instructions.

persona validate

Check workspace for missing or malformed persona files.

python3 scripts/persona-validate.py --workspace ~/.openclaw/workspace

# With config validation
python3 scripts/persona-validate.py --workspace ~/.openclaw/workspace --config ~/.openclaw/openclaw.json

# JSON output for scripting
python3 scripts/persona-validate.py --workspace ~/.openclaw/workspace --json

persona diff

Compare current workspace files against what the generator would produce.

python3 scripts/persona-diff.py --workspace ~/.openclaw/workspace --config ~/.openclaw/openclaw.json

persona list

List all personas across workspaces.

python3 scripts/persona-list.py

# JSON output
python3 scripts/persona-list.py --json

persona fleet

Show a fleet view of all agents across machines.

python3 scripts/persona-fleet.py

# JSON output
python3 scripts/persona-fleet.py --json

voice audition

Audition multiple voices for a provider.

# List ElevenLabs voices
python3 scripts/voice-setup.py --audition --provider elevenlabs

# Filter by gender
python3 scripts/voice-setup.py --audition --provider elevenlabs --gender female

# List built-in voices
python3 scripts/voice-setup.py --audition --provider builtin

Personality Blending

Create hybrid personalities by blending multiple archetypes with weighted percentages.

# During wizard: choose "Blend with a second archetype" option
openclaw persona create

# Programmatically:
cat > blend-profile.json << 'EOF'
{
    "name": "Muse",
    "emoji": "🎨",
    "archetypes": [
        {"name": "companion", "weight": 0.7},
        {"name": "creative", "weight": 0.3}
    ]
}
EOF
python3 scripts/generate-soul.py --input blend-profile.json --output SOUL.md

Blending merges:

  • Traits: Combined from both profiles, weighted by presence
  • Brevity: Weighted average (e.g., 0.7×3 + 0.3×4 = 3.3 → 3)
  • Emotional depth: Weighted average mapped to levels
  • Vibe summaries: Concatenated from both profiles
  • Platform notes: Combined, deduplicated

Community Templates

Pre-built personality profiles for common use cases in assets/community-templates/:

TemplateDescription
financial-advisorInformed, cautious, data-driven financial guide
fitness-coachMotivating, knowledgeable, encouraging workout companion
kids-tutorPatient, fun, age-appropriate learning buddy (ages 6-12)
creative-writerLyrical, perceptive collaborative writing partner
sales-repPersuasive, empathetic, goal-oriented sales professional
therapistEmpathetic, boundaried emotional wellness companion
gaming-buddyEnthusiastic, strategic, competitive gaming partner
executive-assistantAnticipatory, discreet, ultra-efficient EA

Use community templates directly or blend them with built-in archetypes:

python3 scripts/generate-soul.py --archetype therapist --name "Sage" --emoji "🧘"

File Structure

ai-persona-engine/
├── SKILL.md                     # This file
├── DESIGN.md                    # Full design document
├── scripts/
│   ├── persona-create.sh        # Setup wizard (v2: blending, --dry-run, --workspace)
│   ├── persona-update.sh        # Field updater (v2: safe update pipeline)
│   ├── persona-export.sh        # Bundle exporter (v2: cross-workspace)
│   ├── persona-import.sh        # Bundle importer (v2: cross-workspace)
│   ├── persona-preview.py       # [NEW] Sample conversation generator
│   ├── persona-migrate.py       # [NEW] Reverse-engineer config from workspace
│   ├── persona-validate.py      # [NEW] Workspace validation
│   ├── persona-diff.py          # [NEW] Diff workspace vs generated
│   ├── persona-list.py          # [NEW] List all personas across workspaces
│   ├── persona-fleet.py         # [NEW] Fleet view of all agents
│   ├── generate-soul.py         # SOUL.md generator (v2: blending support)
│   ├── generate-user.py         # USER.md generator
│   ├── generate-identity.py     # IDENTITY.md generator
│   ├── generate-memory.py       # MEMORY.md + memory/ scaffolding
│   ├── voice-setup.py           # Voice provider config (v2: --audition)
│   ├── image-setup.py           # Image provider config + reference image
│   └── lib/
│       ├── providers.py         # Provider abstraction layer
│       ├── templates.py         # Template engine + personality blending
│       ├── config.py            # openclaw.json helpers
│       └── prompts.py           # Wizard prompt definitions
├── references/
│   ├── voice-providers.md       # Voice provider setup guide
│   ├── image-providers.md       # Image provider setup guide
│   ├── personality-archetypes.md # Archetype reference
│   ├── soul-writing-guide.md    # SOUL.md writing tips
│   ├── config-schema.md         # openclaw.json schema reference
│   └── migration-guide.md       # [NEW] Migration guide for existing users
├── assets/
│   ├── personality-profiles/    # JSON presets per archetype (5 built-in)
│   ├── community-templates/     # [NEW] 8 community personality profiles
│   └── templates/               # Handlebars templates for generated files
└── tests/
    ├── test-generators.py       # 77 unit tests (v2: blending, migration, etc.)
    ├── test-wizard.sh
    └── test-export-import.sh

Configuration

The persona engine adds a persona section to openclaw.json. Key fields:

FieldTypeDescription
persona.namestringAgent's display name
persona.emojistringAgent's emoji identifier
persona.identityobjectCreature type, vibe, nickname
persona.voiceobjectVoice provider and settings
persona.imageobjectImage provider and canonical look
persona.personalityobjectArchetype, traits, communication style
persona.personality.archetypesarray[NEW] Blend sources with weights
persona.memoryobjectMemory capture and curation settings

See references/config-schema.md for the full schema.


Dependencies

Required:

  • OpenClaw (any version with skill support)
  • Python 3.9+
  • Node.js 22+

Optional (based on chosen providers):

  • ElevenLabs API key (voice)
  • Gemini API key (image generation)
  • xAI API key (Grok TTS / Grok Imagine)
  • ffmpeg (audio format conversion for WhatsApp voice messages)

Further Reading

平台分布

OpenClaw

83.76%
按下载量换算1,154

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills