vllm-project
- 31 skills
- 0 followers
- 2 months ago last updated
- ▌ Add Recipe · vllm-projectUse when the user asks to add, update, validate, or publish a vLLM model recipe in this repo. Covers canonical Hugging Face identity, source/image requirements, checkpoint staging, hardware evidence, YAML-to-UI relationships, command validation, JSON generation, and staging the reviewed files.
- ▌ Pr Review · vllm-projectUse 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.
- ▌ Executor Architecture Review · vllm-projectUse when planning or reviewing agentic-api changes that affect executor streaming, upstream SSE handling, output-item lifecycle, stream delivery, backpressure, or worker placement.
- ▌ Vllm Nightly Perf · vllm-project bundleInstall, migrate, operate, or diagnose the portable vLLM nightly perf-eval automation. Use when moving the nightly Buildkite trigger and Slack regression report to another Linux host, checking their systemd timers, running either job safely, or investigating missing or duplicate nightly perf reports.
- ▌ Vllm Main CI Triage · vllm-project bundleDiagnose hard vLLM main-CI failures, open narrow source-fix PRs when authorized, audit the culprit PR's exact job coverage, and record confirmed test-selection leaks. Use when monitoring the vLLM main-CI dashboard, investigating a failed Buildkite main or daily job, deciding whether a failure is source-related or environmental, or updating the ci-infra selection-leak corpus.
- ▌ Vllm Fast CI Failure Alert · vllm-project bundleInstall, migrate, operate, or diagnose the portable vLLM fast CI failure Slack alert. Use when moving the Databricks-backed short-job alert to another systemd Linux host, checking its 15-minute timer, previewing matching Buildkite CI jobs, testing Slack delivery with explicit approval, or investigating missing or duplicate alerts.
- ▌ Maintainer Ops · vllm-projectUse for vLLM Semantic Router issue intake, PR decisions, releases, and reviewed GitHub mutations performed on behalf of maintainers.
- ▌ Routing Calibration · vllm-projectUse when calibrating or validating a maintained routing recipe against live model backends, including probe manifests and evidence reports.
- ▌ Router Contract Change · vllm-projectUse when changing a request-visible router contract across configuration, signals, decisions, algorithms, or recipe-scoped plugins in vLLM Semantic Router.
- ▌ Openclaw Vsr Bridge · vllm-projectInstall vLLM Semantic Router in agent-safe mode, import supported OpenClaw model providers into canonical VSR config, and rewrite OpenClaw to target VSR.
- ▌ Run E2e · vllm-project bundleUse when the user asks to run, validate, or diagnose the AFD plugin's DeepSeek-V2-Lite GPU/NPU, Qwen3 MoE GPU, or Qwen3.6 MoE CUDA end-to-end tests through the Qwen3.5/3.6 adapter family, including PR-gate E2E, GSM8K-7 accuracy, graph, eager, DBO, or 2A2F scenarios.
- ▌ Review Pr · vllm-project bundleReview pull requests and local branches for vllm-project/afd-plugin with a frozen snapshot, module-design ownership, feature overlays, patch-contract enforcement, targeted validation, and concise evidence-backed findings. Use for default, detailed, or repeat reviews; checking correctness, NPU/GPU compatibility, vLLM 0.26.0 patch discipline, connectors, distributed topology, workers and model runners, tests, benchmarks, or model-family additions; and identifying or explicitly requesting code-owner reviewers. Use run-e2e instead to execute E2E suites, and AGENTS.md for authoring rules.
- ▌ Adapt Model · vllm-project bundleGuide feasibility analysis, minimal implementation, review, and evidence validation for a new AFD model on GPU, NPU, or both. Use when Codex must adapt or review a new vLLM model, establish its native and AFD execution contract, add model-specific tests, or validate model support. Do not use for ordinary model explanations, existing-only E2E execution, generic bug fixes, or vLLM upgrades.
- ▌ Upgrade Npu · vllm-project bundleUpgrade and align the AFD Plugin NPU backend across pinned vLLM and vLLM-Ascend tags or commits. Use when Codex must analyze or implement an NPU runtime upgrade, rebase compatibility patches, adapt NPU models/workers/model runners/connectors, resolve version-driven NPU regressions, or complete the full NPU E2E and accuracy qualification for a new runtime pair. Do not use for GPU-only upgrades, model-only adaptation, ordinary NPU bug fixes unrelated to an upstream upgrade, or E2E-only execution.
- ▌ Upgrade Gpu Version · vllm-project bundleUpgrade and align the AFD Plugin GPU backend across exact pinned vLLM revisions and CUDA runtime/toolchain environments. Use when Codex must plan, audit, implement, review, or finally validate a GPU vLLM version upgrade; rebase compatibility patches; adapt GPU models, workers, model runners, connectors, CUDA Graph, DBO, TP, profiling, packaging, or isolation contracts; or resolve GPU regressions caused by an upstream version change. Do not use for NPU-only upgrades, model-only adaptation, ordinary GPU bugs unrelated to an upstream upgrade, or E2E-only execution.
- ▌ Add Tests And CI · vllm-projectGuide for adding or updating vime tests and CI wiring. Use when tasks require new test cases, CI registration, test matrix updates, or workflow template changes.
- ▌ Add Dynamic Filter · vllm-projectGuide for adding dynamic/filter hooks in vime rollout pipeline. Use when user wants sample-group selection during rollout, buffer filtering before training, or per-sample masking/processing hooks.
- ▌ Add Reward Function · vllm-projectGuide for adding a custom reward function in vime and wiring it through --custom-rm-path (and optional reward post-processing). Use when user wants new reward logic, remote/service reward integration, or task-specific reward shaping.
- ▌ Add Rollout Function · vllm-projectGuide for adding a new rollout function in vime and wiring it through --rollout-function-path. Use when user wants to implement custom rollout data generation logic, custom train/eval rollout outputs, or migrate from the default vLLM rollout path.
- ▌ Add Eval Dataset Config · vllm-projectGuide for adding and validating evaluation dataset configuration in vime. Use when user wants to configure eval datasets via --eval-config or --eval-prompt-data, add per-dataset overrides, or customize evaluation rollout behavior.
- ▌ Vime Code Review Preferences · vllm-projectUse when reviewing or editing vime code, especially refactors around helper APIs, branch selection, argument validation, or recurring reviewer preferences about avoiding unnecessary wrappers and making control flow self-explanatory.
- ▌ Vllm Deploy K8S · vllm-project bundleDeploy vLLM to Kubernetes (K8s) with GPU support, health probes, and OpenAI-compatible API endpoint. Use this skill whenever the user wants to deploy, run, or serve vLLM on a Kubernetes cluster, including creating deployments, services, checking existing deployments, or managing vLLM on K8s.
- ▌ Vllm Bench Serve · vllm-projectBenchmark vLLM or OpenAI-compatible serving endpoints using vllm bench serve. Supports multiple datasets (random, sharegpt, sonnet, HF), backends (openai, openai-chat, vllm-pooling, embeddings), throughput/latency testing with request-rate control, and result saving. Use when benchmarking LLM serving performance, measuring TTFT/TPOT, or load testing inference APIs.
- ▌ Vllm Deploy Docker · vllm-projectDeploy vLLM using Docker (pre-built images or build-from-source) with NVIDIA GPU support and run the OpenAI-compatible server.
- ▌ Vllm Deploy Simple · vllm-project bundleQuick install and deploy vLLM, start serving with a simple LLM, and test OpenAI API.
- ▌ Vllm Prefix Cache Bench · vllm-projectThis is a skill for benchmarking the efficiency of automatic prefix caching in vLLM using fixed prompts, real-world datasets, or synthetic prefix/suffix patterns. Use when the user asks to benchmark prefix caching hit rate, caching efficiency, or repeated-prompt performance in vLLM.
- ▌ Vllm Bench Random Synthetic · vllm-projectRun vLLM performance benchmark using synthetic random data to measure throughput, TTFT (Time to First Token), TPOT (Time per Output Token), and other key performance metrics. Use when the user wants to quickly test vLLM serving performance without downloading external datasets.
- ▌ Guidellm Weekly Summary · vllm-project bundleSummarize 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.
- ▌ Kernel Microbenchmark · vllm-project bundleBuild, debug, and interpret vLLM GPU kernel microbenchmarks for CUDA, Triton, and CuteDSL, including CUPTI timing, correctness checks, generated-code inspection, multi-GPU measurements, and SOL sanity checks.
- ▌ Triton Kernel Writing · vllm-project bundleWrite or review Triton kernels for vLLM, with practical guidance for generated-code inspection, launch grids, indexing, specialization, tuning, and representative performance validation.
- ▌ CI Fails Buildkite · vllm-projectFetch and diagnose vLLM Buildkite CI failure logs. Use when investigating failing CI jobs on a PR or build, when the user pastes a buildkite.com URL, or asks to fetch/diagnose CI logs.