fine-tuning skills

unslothai/unsloth

Skill Claude Code

Create a reusable skill when the user wants Unsloth to learn a repeatable workflow or set of instructions.

aiming-lab/MetaClaw

Skill Claude CodeCodex

Use this skill before any data analysis, transformation, or modeling. Always inspect and validate the data before drawing conclusions or writing transformations.

NVIDIA-NeMo/Nemotron

Skill Claude CodeCodex

Use when planning, debugging, tuning, evaluating, exporting, or deploying public Nemotron embed/rerank retrieval recipes.

oxbshw/LLM-Agents-Ecosystem-Handbook

Skill Claude CodeCodex

Use before opening a PR to audit the changes for stale comments, unused imports, missing tests, and inconsistencies with neighboring code.

zorost/AI-Engineering-Lab

Skill Claude CodeCodex

Wrap an agent loop with step limits, cost caps, human approval gates, and a full trace. Use whenever building or reviewing any tool-calling agent before it touches real systems.

TYH-labs/unsloth-buddy

Skill Claude CodeCodex

This skill should be used when users want to fine-tune language models or perform reinforcement learning (SFT, DPO, GRPO, ORPO, KTO, SimPO) using the highly optimized Unsloth library. Covers environment setup, LoRA patching, VRAM optimization, vision/multimodal fine-tuning, TTS, embedding training, and…

ai4protein/VenusFactory2

Skill Claude CodeCodex

Query NCBI Gene via E-utilities/Datasets API. Search by symbol/ID, retrieve gene info (RefSeqs, GO, locations, phenotypes), batch lookups, for gene annotation and functional analysis.

huaweicloud/huaweicloud-skills

Skill Claude CodeCodex

Manage CloudRobo model training tasks and simulation reinforcement learning (SimRL) tasks — create pretrain (TRAINFROMSCRATCH) and finetune (MODELTUNING) tasks with FFT/SFT/LORA/QLORA/DEEPSPEED methods; manage the full task lifecycle (create/read/update/delete/stop/restart/resume/draft); save and resubmit draft…

gvkhosla/pi-tinker

Skill Claude CodeCodex

Sample, evaluate, and post-train Inkling and Inkling-Small from Pi. Use when the user mentions Inkling, tml-renderers, TMLv0, thinking effort, Inkling-Small, or Inkling audio/images. Load before writing Inkling training or eval code.

lamenting-hawthorn/SkillLoop

Skill Claude CodeCodex

The /goal pattern: cron jobs that run until an objective condition is met, checked by an independent model (maker/checker split). Source: loop engineering framework by 0xCodez / Anthropic evaluator-optimizer pattern.

elimorals/jw-agent-toolkit

Skill Claude CodeCodex

Answer doctrinal questions with verified citations from jw.org. Use when the user asks "¿qué dice la Biblia / Watch Tower sobre X?", "¿qué creemos sobre Y?", "respóndele a este argumento", or any topic that touches on JW doctrine (Trinity, soul, last days, blood, holidays, etc.).

artvandelay/api-vs-selfhost-skill

Skill Claude CodeCodex

Decide API-vs-self-host LLM economics and fine-tuning ROI from any user context (code, PRDs, traffic logs, billing screenshots). Fetches live GPU prices from Runpod/Lambda/Modal, API prices from models.dev or vendor pages, and quality rank from lmarena.ai, then calls a deterministic local Python script for VRAM…

walkowicz19/klayer

Skill Claude CodeCodex

Automatically use klayer for grounding, codebase indexing, session memory, and episode logging during software development tasks unless the user explicitly opts out.

Casius999/fine-tuning-os

Skill Claude CodeCodex

Use this skill when conducting a professional LLM fine-tuning engagement in Zero-Data mode via the MCP server fine-tuning-os. Triggers on: prestation de fine-tuning LLM, livraison de modèle affiné, pipeline Zero-Data, orchestration des 64 outils fine-tuning-os, cycle de vie entraînement (préparation / données…