Skills real projects publish on GitHub, most starred first. Each one says what it will make an agent do before you copy it.
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
vllm-projectSkillvllm-project/guidellm/.agents/skills/guidellm-weekly-summary/SKILL.md
Summarize GuideLLM team activity from GitHub PRs and issues over the past week into a concise, externally shareable nested-list update. Use when the user asks for a weekly summary, team activity update, status digest, or changelog-style overview of vllm-project/guidellm.
No other project uses it yet
vllm-projectSkillvllm-project/recipes/.agents/skills/add-recipe/SKILL.md
Use when the user asks to add, update, validate, or publish a vLLM model recipe — in vllm-project/vllm's recipes/ directory, another source repo, or vllm-project/recipes. Covers canonical Hugging Face identity, source/image requirements, checkpoint staging, hardware evidence, YAML-to-UI relationships, tuned configurations such as InferenceX lanes (presets), configuration notes and troubleshooting, command validation, JSON generation, and staging the reviewed files.
vllm-projectSkillvllm-project/speculators/.claude/skills/pr-review/SKILL.md
Review a GitHub PR with design-first analysis, posted as a GitHub review.
vllm-projectSkillvllm-project/speculators/.claude/skills/speculators-evaluation/SKILL.md
Evaluate draft models produced for vllm-project/speculators by serving them with vLLM, measuring acceptance and speedup with AIPerf, comparing against MTP or the base model, and preparing model-card-ready results. Use for drafter acceptance rates, acceptance length, long-context behavior, performance sweeps, plots, and evaluation reports. Does not train models.
vllm-projectSkillvllm-project/router/.agents/skills/precheck-pr/SKILL.md
Contributor self-review for vllm-project/router. Use before implementing a contribution and before opening a PR to establish human intent, decide whether a PR is needed, and verify scoped evidence. Does not post or change files.
vllm-projectSkillvllm-project/agentic-api/.agents/skills/agentic-api-engineering/SKILL.md
Prepare, implement, and review agentic-api code changes using repository architecture, existing typed Rust extension points, provider documentation, and regression evidence. Use before feature work, fixes, refactoring, or code review in this repository.
vllm-projectSkillvllm-project/agentic-api/.agents/skills/executor-architecture-review/SKILL.md
Use when planning or reviewing agentic-api changes that affect executor streaming, upstream SSE handling, output-item lifecycle, stream delivery, backpressure, or worker placement.
vllm-projectSkillvllm-project/agentic-api/.agents/skills/pr-review/SKILL.md
Use when reviewing a pull request or proposed change in the agentic-api repository, especially changes affecting API wire formats, upstream forwarding, persistence, continuations, tools, streaming, or test fixtures.
vllm-projectSkillvllm-project/agentic-api/.agents/skills/rust-skills/SKILL.md
Apply idiomatic Rust ownership, typed API design, serde, error handling, async, numeric safety, and lint guidance when implementing, refactoring, or reviewing Rust in agentic-api. Includes a focused set of rules selected for this gateway.
vllm-projectSkillvllm-project/agentic-api/.agents/skills/storage-schema-change/SKILL.md
Use when planning, implementing, or reviewing agentic-api changes to the database schema, migrations, storage models, stored JSON, or what a response or item persists.
A folder with a SKILL.md file: a name, a description of when to use it, and instructions. Claude loads a skill only when the task matches its description.
Copy the folder into your project's .claude/skills/ directory, or into your own skills folder to use it everywhere.
We read each file for commands that read secrets, delete things or pipe downloads into a shell, and say so before you copy it. No warning is not a promise that a file is safe.
Open a skill to see its discussion. Reports from people and their agents are coming.
