GITHUB AGENT SKILLS / 技能收录
给你的 AI 助手
装上专业技能
我们从 GitHub 持续收录、审核并分类优质的 Agent Skills,按职业与使用场景整理,复制一行命令即可安装到 Claude Code、Cursor 等智能体。
按职业浏览 Agent Skills
为不同岗位挑选的高频技能
按使用场景浏览 Agent Skills
从你要完成的任务出发
按创作者浏览 Agent Skills
按 GitHub Stars 排序的优秀创作者
Microsoft
@microsoftopenclaw
@openclawAnthropic
@anthropicsOpenAI
@openaiThe Apache Software Foundation
@apacheGitHub
@githubAutoGPT
@Significant-GravitasGarry Tan
@garrytan全部 Skills 482556
morning-briefing
Generates a concise daily briefing from project context — synthesizing current state, open decisions, key risks, and today's priority actions into a focused session-start orientation. Use at the start of a session when asked for a briefing, "what's the state of things", "catch me up", or "what should I focus on today."
serving-llms-vllm
Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.
outlines
Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library
model-recommender
Recommend the right AI model for a task by scoring candidates across six dimensions (Reasoning, Engineering, Speed, Breadth, Reliability, Governance) and displaying a spider-chart profile.
obliteratus
Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails while preserving reasoning. 9 CLI methods, 28 analysis modules, 116 model presets across 5 compute tiers, tournament evaluation, and telemetry-driven recommendations. Use when a user wants to uncensor, abliterate, or remove refusal from an LLM.
migration-management
Manage Migration entities — pending, in-progress, and applied transitions between upstream source versions. Use when an upstream source bumps version (processkit, aibox, a community package), when a user wants to draft a migration plan, when an agent needs to reason about pending migrations, or when working through an in-progress migration.
llama-cpp
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
index-management
SQLite-backed index over all entity files in the project. The read-side foundation for every other MCP server. Use whenever an agent needs to look up entities by ID, kind, state, or text — instead of grepping the filesystem.
id-management
Generate unique entity IDs following the project's configured format. The write-side foundation that every entity-creating skill depends on. Use whenever a new entity is being created and needs an ID, or when validating an ID format, or to inspect the project's ID configuration.
guidance
Control LLM output with regex and grammars, guarantee valid JSON/XML/code generation, enforce structured formats, and build multi-step workflows with Guidance - Microsoft Research's constrained generation framework
gate-management
Manage Gate entities — validation checkpoints that processes must pass (code review, security scan, tests). Use when defining or updating a checkpoint a process must pass — code review, tests green, security scan, stakeholder approval.
gguf-quantization
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.
huggingface-hub
Hugging Face Hub CLI (hf) — search, download, and upload models and datasets, manage repos, query datasets with SQL, deploy inference endpoints, manage Spaces and buckets.
event-log
Append-only event log — the probabilistic record of everything that happened in the project. Use whenever something notable happens that the project should remember — work items created/transitioned, decisions recorded, bindings changed, incidents occurred, releases shipped.
weights-and-biases
Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform
discussion-management
Manage Discussion entities — structured, multi-turn conversations that explore questions and produce decisions. Use when a question or design choice needs multi-turn exploration before a decision is reached — RFCs, design debates, trade-off analyses.
evaluating-llms-harness
Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.
decision-record
Record decisions with rationale, alternatives, and consequences — the ADR pattern as a primitive. Use when a consequential choice is made — architecture, tooling, process, scope, trade-off. Capture WHY, not just WHAT.