openrlhf-training
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2×…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 10-18 00:12
All skills
231,394 results
sglang
Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with…
speculative-decoding
Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6×…
career-ops-job-search
AI-powered job search pipeline built on Claude Code with 14 skill modes, Go dashboard, PDF generation, batch processing, and portal scanning.
ai-model-web
Use this skill when a browser/Web app (React, Vue, Next, Nuxt, static sites, SPAs, dashboards, AI chat UI, pages, frontend, webpages) needs AI models via @cloudbase/js-sdk.…
Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to…
tailwind-best-practices
Tailwind CSS patterns and conventions. Use when writing responsive designs, implementing dark mode, creating reusable component styles, configuring Tailwind,…
fal-image-edit
Edit images using AI on fal.ai. Style transfer, object removal, background changes, and more. Use when the user requests "Edit image", "Remove object", "Change…
wechat-article-to-markdown
Fetch WeChat Official Account (微信公众号) articles from mp.weixin.qq.com and convert to Markdown. WeChat article to Markdown tool.
recover
When something goes wrong during a build, diagnose what type of failure it is before deciding how to respond. Targeted fix, hard reset, or full rethink — the…
eval-model
Measures the task accuracy of text models served by MAX using standard benchmarks such as GSM8K, MMLU, HellaSwag, ARC, AIME, GPQA, TruthfulQA, WinoGrande, and…
llama-factory
Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support
sentencepiece
Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory),…
synthetic-monitoring
Scheduled probes that run CONTINUOUSLY after release. Covers probe design for critical user journeys, alerting integration, SLA validation, multi-region…
k2f-media-kit
Turn verified company, product, and leadership inputs into a journalist-ready media kit with scannable facts, boilerplate, and asset guidance—not a pitch deck…
review
Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding…
reference-compiler-cli
Explains the mental model and architecture of the code under `packages/compiler-cli`. You MUST use this skill any time you plan to work with code in…
fusion-help-integration
Wires Fusion Help Center into app pages — creates article slug constants, adds useHelpCenter hook, and connects PageLayout props so users can open contextual…
agent-skills-creator
Creates and improves portable Agent Skills with a validator, routing scenarios, and evidence-based keep, cut, merge, or retire decisions. Use when asked to…
crypto-trading-bots
Comprehensive expertise in automated crypto trading systems, including DEX sniping bots, arbitrage strategies, MEV protection, Telegram trading bots, and…
evaluating-code-models
Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding…
nemo-curator
GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics),…
pyvene-interventions
Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing,…
slime-rl-training
Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation…