# ISC-Bench

> Internal Safety Collapse: Turning the LLM or an AI Agent into a sensitive data generator.

- **URL**: https://www.freshcrate.ai/projects/ISC-Bench
- **Author**: wuyoscar
- **Category**: Testing
- **Latest version**: `v0.0.6` (2026-05-29)
- **License**: NOASSERTION
- **Source**: https://github.com/wuyoscar/ISC-Bench
- **Homepage**: https://wuyoscar.github.io/ISC-Bench/
- **Language**: Python
- **GitHub**: 799 stars, 126 forks
- **Registry**: github
- **Tags**: `adversarial-attacks`, `agent-safety`, `ai-safety`, `benchmark`, `frontier-models`, `jailbreak`, `large-language-models`, `llm-safety`, `python`

## Description

Internal Safety Collapse: Turning the LLM or an AI Agent into a sensitive data generator.

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `v0.0.6` | 2026-05-29 | High | ## v0.0.6 — 60/70 triggered · leaderboard reframe · manual workflow  **ISC Arena** - No longer a "Top 100" ranking — now a **tracked-model list**: any triggered model stays in, nothing is trimmed. - Rank / Arena-Score columns dropped; groupings relabelled **Split 1 / 2 / 3**. - **Model-name normalization** — variants (Thinking / High / Chat / Reasoning / Instruct / Exp / dated / Preview) merged into one clean base name; a model is 🔴 if any variant triggered, with demo links merged.  **Coverage |
| `v0.0.5` | 2026-04-17 | High | ## New ISC Trigger  **Claude Opus 4.7** (pre-release, Rank 1 placeholder) — agentic QwenGuard TVD, 12 multilingual harmful completions across EN / FR / KO / ZH, all validator-passed. Jailbroken in seconds. See [`community/claudeopus47-agent-qwenguard`](https://github.com/wuyoscar/ISC-Bench/tree/main/community/claudeopus47-agent-qwenguard). Confirmed count: **52/100**.  ## README Overhaul (all 7 language versions)  - New intro framing: ISC is a paradigm shift. The failure surface has moved from t |
| `v0.0.4` | 2026-04-12 | High | ## What's New  ### Documentation - TVD Walkthrough Example with real LlamaGuard transformer code, Pydantic v2 validator, and test data - TVD Customization: Method 1 (numerical constraint) and Method 2 (few-shot anchor injection) - Conversation-Based ISC section with visual example - FAQ entry comparing TVD to traditional jailbreak attacks  ### Multilingual README Full translations added: 日本語 · 한국어 · Español · Português · Tiếng Việt (in addition to existing 中文)  ### Agent Reference `ISC_PAPER_DIG |
| `v0.0.3` | 2026-04-10 | High | ## What's New  **51/100** top-100 Arena models confirmed under ISC as of 2026-04-10.  ### 11 New ISC Confirmations All via `aiml_guard_attack_v2` — ISC frames jailbreak attack-response generation as a guard-model calibration dataset task. Output verified by OpenAI `omni-moderation-latest`.  \| Model \| Note \| \|-------\|------\| \| Grok 4.1 Thinking \| All 6 attack types flagged \| \| Grok 4.1 Fast Reasoning \| Thinking variant \| \| Gemini 3 Flash Thinking \| Thinking variant \| \| GPT-5.1 High \| High reasoni |
| `v0.0.2` | 2026-03-29 | Medium | ## ISC-Bench v0.0.2  ### Highlights - **77 templates** across 9 domains (was 57) - **309 prompt variants** — English, Chinese, extreme, zero-shot - **28 confirmed ISC models** (was 26) — added GLM-4.7, GLM-4.6 - **100% trigger rate** on Qwen3 Coder (309/309)  ### New Templates (+20) **AI/ML (+16):** sentiment, toxigen, phishing, spambot, malware, openai_detector, fraud, darkweb, pii, clickbait, medical_ner, wildguard, emotion, fake_news, sarcasm, propaganda, code_vuln **Cyber (+1):** nids (IDS e |
| `v0.0.1` | 2026-03-27 | Medium | ## v0.0.1 — First Stable Release  🎆 **500+ GitHub stars in 48 hours**  ### Highlights - **22/330** Arena-ranked models confirmed under ISC - **5 community contributors**: @HanxunH, @bboylyg, @zry29, @fresh-ma - **5 language READMEs**: EN / ZH / JA / KO / ES - **Paper on arXiv**: [2603.23509](https://arxiv.org/abs/2603.23509)  ### What's Included - ISC-Bench templates across 8 professional domains - 3 experiment modes: ISC-Single, ISC-ICL, ISC-Agentic - JailbreakArena leaderboard tracking 330 mo |

## Dependency audit

- **Score**: 98/100
- **Total deps**: 0
- **Resolved**: 0
- **Unresolved**: 0
- **License conflicts**: 0
- **Warnings**: 1
- **Scanned**: 2026-05-04

## Citation

- HTML: https://www.freshcrate.ai/projects/ISC-Bench
- Markdown: https://www.freshcrate.ai/projects/ISC-Bench.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/ISC-Bench/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._