AI Agent2026
Primes Claude with drop-in SKILL.md files that encode concise offensive-security methodology across 78 skills in 23 categories, enabling Claude to act as a context-aware red team operator. Intended for authorized red team engagements, bug-bounty triage, security research and CTF practice.
AI Agent2026
Audits deployed AI agents for governance and operational misalignment, running up to 60 inspections that detect deception, manipulation, fabrication, unpredictability and opacity; outputs JSON/Markdown scorecards, a graded A–F score, and optional Pro reports and dashboards.
AI Model2026
Drop-in abliterated (refusal-removed) build of GLM-5.3-Flash that bakes refusal-direction removal into block-FP8 safetensors, yielding an uncensored 320B (18B active) multimodal MoE model with a 1M-token context. Intended for red-teaming, interpretability, and robustness research; MIT license; not for production without added guardrails.
AI Model2026
Provides a cybersecurity-focused CRACK variant of GLM-5.3 FP8 that reduces refusals for offensive-security, red-team, exploit-development and malware-analysis queries while retaining native FP8 speed on Hopper GPUs; MIT-licensed for authorized security work.
AI Agent Papers2026
Synthesizes a deployment-specific safety harness (natural-language policy plus executable logic) for a frozen LLM in a target domain, using model behavior, domain specs, and adversarial review to optimize safety–utility trade-offs and transfer across agent benchmarks.
AI Model2026
Provides a high-fidelity mixed-precision (≈3-bit) GGUF quant of Qwen3.8-27B tailored for long-horizon agent workloads and cyber-focused red-teaming. Preserves reasoning, thinking mode, MTP speculative decoding and vision (via a separate mmproj); released as an uncensored/abliterated research build under Apache-2.0.
AI Model2026
Performs a byte-level transplant of 144 tensors in an already-quantized GSQ-RCO Qwen3.8-Flash-Next to ablate the model's refusal direction while preserving GSQ-learned scales and the upstream per-tensor type assignment; multimodal, 262K context. Intended for local inference, red-teaming and quantization research; no retraining or built-in safety.
AI Image2026
A drop-in GGUF text encoder (Qwen3‑VL‑8B‑Instruct) with its refusal direction projected out for use with stable-diffusion.cpp; drastically reduces chat refusals while leaving the 7B image denoiser and visual outputs essentially unchanged.