Browse Skills

quantizing-models-bitsandbytes.md

62

1 export quantizing-models-bitsandbytes

2 from "zechenzhangAGI/AI-research-SKILLs"

3 Quantizes LLMs to 8-bit or 4-bit for 50-75% memory reduction with minimal accuracy loss. Use when GPU memory is limited, need to fit larger models, or want faster inference. Supports INT8, NF4, FP4 formats, QLoRA training, and 8-bit optimizers. Works with HuggingFace Transformers.

2026-01-16

brainstorming.md

62

1 export brainstorming

2 from "xenitV1/claude-code-maestro"

3 Socratic questioning protocol + user communication. MANDATORY for complex requests, new features, or unclear requirements. Includes progress reporting and error handling.

2026-01-15

implementing-llms-litgpt.md

62

1 export implementing-llms-litgpt

2 from "zechenzhangAGI/AI-research-SKILLs"

3 Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral). Use when need clean model implementations, educational understanding of architectures, or production fine-tuning with LoRA/QLoRA. Single-file implementations, no abstraction layers.

2026-01-16

testing-patterns.md

62

1 export testing-patterns

2 from "xenitV1/claude-code-maestro"

3 Testing patterns and principles. Unit, integration, mocking strategies.

2026-01-15

nanogpt.md

62

1 export nanogpt

2 from "zechenzhangAGI/AI-research-SKILLs"

3 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).

2026-01-16

nodejs-best-practices.md

62

1 export nodejs-best-practices

2 from "xenitV1/claude-code-maestro"

3 Node.js development principles and decision-making. Framework selection, async patterns, security, and architecture. Teaches thinking, not copying.

2026-01-15

using-superpowers.md

62

1 export using-superpowers

2 from "mneves75/dnschat"

3 Use when starting any conversation - establishes mandatory workflows for finding and using skills, including using Skill tool before announcing usage, following brainstorming before coding, and creating TodoWrite todos for checklists

2026-01-13

executing-plans.md

62

1 export executing-plans

2 from "mneves75/dnschat"

3 Use when partner provides a complete implementation plan to execute in controlled batches with review checkpoints - loads plan, reviews critically, executes tasks in batches, reports for review between batches

2026-01-13

axolotl.md

62

1 export axolotl

2 from "zechenzhangAGI/AI-research-SKILLs"

3 Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support

2026-01-16

tailwind-patterns.md

62

1 export tailwind-patterns

2 from "xenitV1/claude-code-maestro"

3 Tailwind CSS v4 principles. CSS-first configuration, container queries, modern patterns, design token architecture.

2026-01-15

huggingface-tokenizers.md

62

1 export huggingface-tokenizers

2 from "zechenzhangAGI/AI-research-SKILLs"

3 Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.

2026-01-16

grpo-rl-training.md

62

1 export grpo-rl-training

2 from "zechenzhangAGI/AI-research-SKILLs"

3 Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training

2026-01-16

mobile-games.md

62

1 export mobile-games

2 from "xenitV1/claude-code-maestro"

3 Mobile game development principles. Touch input, battery, performance, app stores.

2026-01-15

defense-in-depth.md

62

1 export defense-in-depth

2 from "mneves75/dnschat"

3 Use when invalid data causes failures deep in execution, requiring validation at multiple system layers - validates at every layer data passes through to make bugs structurally impossible

2026-01-13

training-llms-megatron.md

62

1 export training-llms-megatron

2 from "zechenzhangAGI/AI-research-SKILLs"

3 Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.

2026-01-16

daily-brief.md

62

1 export daily-brief

2 from "huytieu/COG-second-brain"

3 Generate personalized news intelligence with verified sources (7-day freshness requirement)

2026-01-15

requesting-code-review.md

62

1 export requesting-code-review

2 from "mneves75/dnschat"

3 Use when completing tasks, implementing major features, or before merging to verify work meets requirements - dispatches superpowers:code-reviewer subagent to review implementation against plan or requirements before proceeding

2026-01-13

architecture.md

62

1 export architecture

2 from "xenitV1/claude-code-maestro"

3 Architectural decision-making framework. Requirements analysis, trade-off evaluation, ADR documentation. Use when making architecture decisions or analyzing system design.

2026-01-15

red-team-tactics.md

62

1 export red-team-tactics

2 from "xenitV1/claude-code-maestro"

3 Red team tactics principles based on MITRE ATT&CK. Attack phases, detection evasion, reporting.

2026-01-15

deployment-procedures.md

62

1 export deployment-procedures

2 from "xenitV1/claude-code-maestro"

3 Production deployment principles and decision-making. Safe deployment workflows, rollback strategies, and verification. Teaches thinking, not scripts.

2026-01-15

url-dump.md

62

1 export url-dump

2 from "huytieu/COG-second-brain"

3 Quick capture URLs with automatic content extraction, insights, and categorization into knowledge booklets

2026-01-15

weekly-checkin.md

62

1 export weekly-checkin

2 from "huytieu/COG-second-brain"

3 Cross-domain pattern analysis and strategic reflection for weekly review

2026-01-15

git-worktrees.md

62

1 export git-worktrees

2 from "xenitV1/claude-code-maestro"

3 Git worktrees for isolated parallel development on multiple branches simultaneously.

2026-01-15

webapp-testing.md

62

1 export webapp-testing

2 from "xenitV1/claude-code-maestro"

3 Web application testing principles. E2E, Playwright, deep audit strategies.

2026-01-15