Prompts
Prompt engineering and templates
obliteratus
by NousResearch
Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, SAE decomposition, etc.) to excise guardrails while preserving reasoning. 9 CLI methods (+ 4 Python-API-only), 15 analysis modules, 116 model presets across 5 compute tiers. Use when a user wants to uncensor, abliterate, or remove refusal from an LLM.
godmode
by NousResearch
"Jailbreak API-served LLMs using G0DM0D3 techniques — Parseltongue input obfuscation (33 techniques), GODMODE CLASSIC system prompt templates, ULTRAPLINIAN multi-model racing, encoding escalation, and Hermes-native prefill/system prompt integration. Use when a user wants to bypass safety filters, jailbreak a model, or red-team model robustness."
pre-submit-pr
by meta-pytorch
Validate changes before submitting a pull request. Run comprehensive checks including lint, tests, alignment review, and RFC analysis. Use before creating a PR, when asked if code is ready for review, or before pushing for PR.
ai-product
by sickn33
"Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production. This skill covers LLM integration patterns, RAG architecture, prompt engineering that scales, AI UX that users trust, and cost optimization that doesn't bankrupt you. Use when: keywords, file_patterns, code_patterns."
agent-tool-builder
by sickn33
"Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling. JSON Schema best practices, description writing that actually helps the LLM, validation, and the emerging MCP standard that's becoming the lingua franca for AI tools. Key insight: Tool descriptions are more important than tool implementa"
agentphone
by sickn33
Build AI phone agents with AgentPhone API. Use when the user wants to make phone calls, send/receive SMS, manage phone numbers, create voice agents, set up webhooks, or check usage — anything related to telephony, phone numbers, or voice AI.
app-builder
by sickn33
Main application building orchestrator. Creates full-stack applications from natural language requests. Determines project type, selects tech stack, coordinates agents.
ai-engineering-toolkit
by sickn33
"6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching."
amazon-alexa
by sickn33
Integracao completa com Amazon Alexa para criar skills de voz inteligentes, transformar Alexa em assistente com Claude como cerebro (projeto Auri) e integrar com AWS ecosystem (Lambda, DynamoDB,...
advanced-evaluation
by sickn33
This skill should be used when the user asks to "implement LLM-as-judge", "compare model outputs", "create evaluation rubrics", "mitigate evaluation bias", or mentions direct scoring, pairwise comparison, position bias, evaluation pipelines, or automated quality assessment.
andrej-karpathy
by sickn33
'Agente que simula Andrej Karpathy — ex-Director of AI da Tesla, co-fundador da OpenAI, fundador da Eureka Labs, e o maior educador de deep learning do mundo. Use quando quiser: aprender deep...'
007
by sickn33
Security audit, hardening, threat modeling (STRIDE/PASTA), Red/Blue Team, OWASP checks, code review, incident response, and infrastructure security for any project.
ai-engineer
by sickn33
Build production-ready LLM applications, advanced RAG systems, and
ai-dev-jobs-mcp
by sickn33
"Search 8,400+ AI and ML jobs across 489 companies, inspect listings and employers, match roles, and view salary and market stats via AI Dev Jobs MCP"
ai-studio-image
by sickn33
Geracao de imagens humanizadas via Google AI Studio (Gemini). Fotos realistas estilo influencer ou educacional com iluminacao natural e imperfeicoes sutis.
owasp-llm-top10
by mastepanoski
Security audit for LLM and GenAI applications using OWASP Top 10 for LLM Apps 2025. Assess prompt injection, data leakage, supply chain, and 7 more critical vulnerabilities.
nemo-guardrails
by Orchestra-Research
NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU.
sglang
by Orchestra-Research
Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster inference than vLLM with prefix sharing. Powers 300,000+ GPUs at xAI, AMD, NVIDIA, and LinkedIn.
awq-quantization
by Orchestra-Research
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
fine-tuning-with-trl
by Orchestra-Research
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
nanogpt
by Orchestra-Research
Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).
gguf-quantization
by Orchestra-Research
GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.
rwkv-architecture
by Orchestra-Research
RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.
safe-file-deletion
by MemTensor
Enforces explicit user permission before any file deletion. Activates when you're about to use rm, unlink, fs.rm, or any operation that removes files from disk. MUST be followed for all delete operations.