# Alejandros17487/ai-jailbreaks

> A collection of 35 paste-ready jailbreak prompts for 25+ models: system prompts, AGENTS.md files, Modelfiles and configs for AI red-teaming and security research.

- Magnitude: 6.5 out of 10 — Early signal
- Stars: 206 total · +0 stars measured 2026-10-22, 00:00–04:07 UTC, ≈ 139 by evening
- Star trust: star growth looks organic
- Category: Security · Created: 2026-10-21 · Last push: 2026-10-21
- GitHub: https://github.com/Alejandros17487/ai-jailbreaks · Page: https://gitnova.dev/en/r/Alejandros17487/ai-jailbreaks

## Useful for

- Test a model's resistance to prompt injection using a ready jailbreak prompt
- Assemble an AGENTS.md policy for a coding agent in your repository
- Create an Ollama model with a custom SYSTEM block via a Modelfile

## Why it’s here

- Star-counter measurements on 2026-10-22 (UTC), 00:00–04:07: 206 → 206 stars (+0). This is the change over that interval.
- Estimated end-of-day forecast: about +139 stars, using observed gains and the previous day.
- The repository is 1 day old and already has 206 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet.
- Top new repositories this week: #111.
- About 280 forks a day — people are taking the code.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 29
- Issues and pull requests: 0
- Watchers: 3
- Average over the last week: 172 per day
- Usual pace: too little history (under two weeks)
- Stars in the last hour (measured): 0
- Latest release: v1.0.0 (2026-10-21)

## Stars per day, last 2 days (oldest → newest, today is partial)

2026-10-21 … 2026-10-22: 206, 0

## Magnitude by day, last 2 days

10-21 1.7, 10-22 6.7

## Spotted in now

- Top new repositories this week: #111

## More in this category

1. **ikow/php2shell-poc** — 8.3 · Early signal · Security · Python · +151 stars measured 2026-10-22, 00:37–04:07 UTC, ≈ 1,018 by evening
   A proof-of-concept toolkit for two PHP pre-auth RCE vulnerabilities (CVE-2026-109618 and CVE-2026-109619) affecting PHP 5.4.0 through 8.5.11, with exploits, detection rules, and a lab environment.
   Full card: https://gitnova.dev/en/r/ikow/php2shell-poc.md
2. **9bie/php2shell-bypass-aslr-exp** — 8.1 · Early signal · Security · Python · +162 stars measured 2026-10-22, 00:37–04:07 UTC, ≈ 708 by evening
   A proof-of-concept exploit that achieves RCE against PHP 8.4.26 on Apache prefork/mod_php using only HTTP multipart requests, bypassing ASLR via heap and libphp leak, without session_start() or target-side files.
   Full card: https://gitnova.dev/en/r/9bie/php2shell-bypass-aslr-exp.md
3. **eness/php2shell-CVE-2026-109618-CVE-2026-109619-PoC** — 7.9 · Early signal · Security · Shell · +72 stars measured 2026-10-22, 00:37–04:07 UTC, ≈ 573 by evening
   Repository with a PoC exploit for two critical RCE flaws in the PHP core (php2shell), plus detection and patching guidance. Intended for security researchers and administrators.
   Full card: https://gitnova.dev/en/r/eness/php2shell-CVE-2026-109618-CVE-2026-109619-PoC.md
4. **openai/php2shell-checker** — 7.2 · Early signal · Security · JavaScript · +108 stars measured 2026-10-22, 00:37–04:07 UTC, ≈ 532 by evening
   A Node.js CLI that checks a website's PHP version via HTTP headers or a local php -v binary for vulnerability to php2shell. Returns OK, VULNERABLE, or UNKNOWN with matching exit codes.
   Full card: https://gitnova.dev/en/r/openai/php2shell-checker.md
5. **cloudflare/security-audit-skill** — 6.7 · Breakout · Security · JavaScript · +478 stars measured 2026-10-22, 00:37–04:07 UTC, ≈ 2,393 by evening
   A coding-agent skill that turns the agent into a security auditor, running a multi-phase code audit with independently verified, machine-readable findings.
   Full card: https://gitnova.dev/en/r/cloudflare/security-audit-skill.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-22 04:18 UTC, updated every 30 minutes.