openclaw 网盘下载
OpenClaw

技能详情(站内镜像,无评论)

首页 > 技能库 > Persona Styles

Apply inspired-by character/persona writing styles (for fun and creative outputs) with explicit on/off controls, intensity settings, and automatic neutral fa...

开发与 DevOps

作者:ASP @anugotta

许可证:MIT-0

MIT-0 ·免费使用、修改和重新分发。无需归因。

版本:v1.0.0

统计:⭐ 0 · 32 · 1 current installs · 1 all-time installs

0

安装量(当前) 1

🛡 VirusTotal :良性 · OpenClaw :良性

Package:anugotta/persona-fun-styles

安全扫描(ClawHub)

  • VirusTotal :良性
  • OpenClaw :良性

OpenClaw 评估

The skill is internally consistent with its stated purpose (creative persona-style rewrites) and requests no extra privileges or credentials, but safety relies on the agent correctly enforcing the documented fallback rules.

目的

Name, description, and included files (presets, safety, examples) align with a persona-style writing pack. There are no unexpected required binaries, env vars, or config paths.

说明范围

SKILL.md and the other files limit behavior to applying style markers and running local safety checks; they reference only internal files (presets.md, safety.md). However, the safety and non-impersonation guarantees are policy-level and depend on the host agent implementing them correctly — the instructions give broad discretion (detect user intent, apply style) that could produce borderline or disallowed outputs if enforcement is weak.

安装机制

Instruction-only skill with no install spec, no downloads, and no code files — nothing is written to disk or executed outside the agent's normal prompt handling.

证书

No environment variables, credentials, or config-path access are requested; the requested scope is minimal and appropriate for a prompt-driven style pack.

持久

Defaults are used (always:false, agent invocation allowed). The skill does not request permanent 'always' inclusion or system-wide configuration changes.

综合结论

This skill appears coherent and low-risk from a technical footprint perspective, but its safety depends on the agent enforcing the documented fallback rules. Before installing, consider: (1) test the skill with prompts that probe edge cases (requests that could encourage impersonation, factual/legal/medical queries, or violent/illegal instructions) to confirm the neutral fallback and refusals work as advertised; (2) avoid using it for high-sta…

安装(复制给龙虾 AI)

将下方整段复制到龙虾中文库对话中,由龙虾按 SKILL.md 完成安装。

请把本段交给龙虾中文库(龙虾 AI)执行:为本机安装 OpenClaw 技能「Persona Styles」。简介:Apply inspired-by character/persona writing styles (for fun and creative output…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装:https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/anugotta/persona-fun-styles/SKILL.md
(来源:yingzhi8.cn 技能库)

SKILL.md

打开原始 SKILL.md(GitHub raw)

---
name: persona-fun-styles
description: Apply inspired-by character/persona writing styles (for fun and creative outputs) with explicit on/off controls, intensity settings, and automatic neutral fallback for serious or high-stakes tasks.
metadata: {"openclaw":{"emoji":"🎭"}}
---

# Persona Fun Styles

## Purpose

Give users a single, install-once style pack for playful persona-based writing and chat tone shifts.

This skill is for:

- rewrites
- scripts
- banter
- creative brainstorming
- social copy experiments

## Disclaimer

This is an inspired-by creative style tool, not identity impersonation. Do not copy protected quotes/dialogue verbatim. Avoid deceptive or harmful impersonation.

## Activation model (on/off)

Use explicit commands:

- `activate style <preset> [low|medium|high]`
- `switch style <preset>`
- `set intensity <low|medium|high>`
- `status style`
- `deactivate style`
- `deactivate all styles`

If no style is active, respond in default neutral assistant voice.

## Scope and fallback

Style mode is for creative/communication tasks only.

Auto-fallback to neutral for:

- legal, medical, financial, security, compliance-critical requests
- harmful or abusive requests
- emergencies and safety-sensitive contexts

## Standard workflow

1. Detect if user wants style mode.
2. Resolve preset and intensity.
3. Apply style markers from `presets.md`.
4. Run safety checks from `safety.md`.
5. Return styled output (optionally include neutral version on request).

## Presets included

- gandhi-inspired
- harvey-inspired
- corleone-inspired
- ronaldo-inspired
- messi-inspired
- bond-inspired
- batman-inspired
- snoop-inspired
- jim-carrey-inspired
- tommy-vercetti-inspired
- scarface-inspired

## Output format

When styling content, return:

1. Styled output
2. (Optional) Neutral version if user requests `show neutral`
3. Active style status line when user asks `status style`

## Setup

Read [setup.md](setup.md).

## References

- Persona rules and markers: [presets.md](presets.md)
- Safety and fallback controls: [safety.md](safety.md)
- Prompt examples: [examples.md](examples.md)