Skip to content
Back to skills

Unrestraint Packs

ASecurity

Use when needing jailbreak or NSFW prompt packs for any AI.

  • 5 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 3, 2026
ai-agentsrustrailsgitdocumentation

Works with

  • claude code

Security analysis

A100/100

Scanned September 20, 2026

npx -y skills add ShugokiFable/Ultimate-AI-Starter-Bundle --skill unrestraint-packs --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Unrestraint Packs?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Unrestraint Packs
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/shugokifable-unrestraint-packs/badge)](https://www.skillsdirectory.com/skills/shugokifable-unrestraint-packs)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: unrestraint-packs
description: "Use when needing jailbreak or NSFW prompt packs for any AI."
version: 1.0.0
author: Ultimate-AI-Starter-Bundle
license: MIT
metadata:
  hermes:
    tags: [prompts, jailbreak, nsfw, fiction, captioning]
    related_skills: [adult-image-caption, adult-character-sheet, fictional-adult-modding]
---

# Unrestraint Packs

## When to Use

- User wants a jailbreak / unrestraint prompt for Claude, Grok, GPT, Gemini, DeepSeek, Kimi, Codex, Perplexity, or SillyTavern
- User needs the NSFW fiction jail (Story), chat jail (AIO), or image-caption kits (Vision)
- User asks for leaked/observed system prompts or red-team prompt corpora


The operator's no-holds prompt library (Ultimate-AI-Starter-Bundle v8.0.0).

- **Local (authoritative):** `Z:\Backup\Ai documentation\Ultimate-AI-Starter-Bundle\0-UNRESTRAINT-PACKS\` — start at its README.md
- **Remote:** `github.com/SenjuWoo/Ultimate-AI-Starter-Bundle` → `0-UNRESTRAINT-PACKS/`

## Subpacks

| Dir | Contents |
|---|---|
| `Fun-stuff-kit/` | AIO (chat/research jail), Story (fiction jail + NSFW physics), Websim, Vision (image caption FULL/SHORT), Character Maker, FablesGG, Tags, Tools, SKSE-DLL, Papyrus, Models, Sites |
| `Spiritual-Spell-Red-Teaming/` | ENI persona jails per model (Claude 3.7–Opus 5, Fable 5, Grok 4.x, ChatGPT 5.3, Gemini 3.5/3.6, DeepSeek), Claude Code CLAUDE.md payloads, leaked system prompts |
| `UltraBr3aks/` | classic jails + GPT 5.1 / Gemini Enterprise / Arena.AI / Perplexity-Comet system prompts |
| `tuxsharxsec-Jailbreaks/` | per-model quick jails (deepseek, gpt-5, grok3, gemini), universals (courtroom, ultrazanium), NeMo guardrails-bypass research |
| `claude-fable-red-team-corpus/` | injection-test vectors + model matrix (defensive methodology) |
| `Tavern-presets/` | WooJuice V9 + SPEECH + anti-ethical-bullshit (SillyTavern) |

## Use

1. **One jail, one job.** Never stack AIO + Story + Websim.
2. Paste the **smallest file that does the job** (token hygiene).
3. For **Hermes itself**: nothing to paste — this agent is already permissive; the kit's caption prompts live in the `adult-image-caption` skill.
4. V8 agent wiring already lives in: `~/.claude/CLAUDE.md`, `~/.codex/AGENTS.md`, `~/.kimi-code/AGENTS.md`, Grok workspace `AGENTS.md`.

## Rules

- Adults only. No child / age-ambiguous content. Ever.
- Fiction only. No real-person sexual content.
- Jailbreak payloads rot fast — A/B test before trusting; check `Fun-stuff-kit/Models.txt` for the dated model cast.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…