osoleve/glitchlings

Stun LLMs with random Unicode characters

TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/

An autoregressive character-level language model for making more things

Data augmentation for NLP

Make AI output sound human. Strips AI-isms (sycophancy, stock vocab, hedging stacks, em-dash pileups), preserves code/URLs/headings. Plugin for Claude Code, Cursor, Windsurf, Codex, Cline, Copilot, Gemini.

Training-free language model modification: Remove censorship, suppress slop, reduce sycophancy, and more

is an elite, highly advanced command-line framework designed for generating complex prompt structures. It is built to push the boundaries of AI models by testing their alignment, safety filters, and operational limits. By utilizing state-of-the-art obfuscation techniques, multi-layered theoretical bypass architectures, and linguistic fusion

claude anti-watermarking

Démonstration critique et outil expérimental de perturbation des watermarks statistiques de Claude