1414 个 · 按相关度排序
Runs a saved development plan by assigning implementation, testing, and review to subagents. Tracks progress and verification evidence in isolated Git working directories, with options to deliver pull requests or continue through merging.
中文简介正在准备中。当前显示英语简介。
Turn implementation plans and specifications into an ordered tasks.md checklist. Organize work around user stories, meaning goals expressed from a user's perspective, with file paths, dependencies, parallel work opportunities, and independent verification criteria.
中文简介正在准备中。当前显示英语简介。
Implement filtering, aggregation, joins, and file processing with Polars for tabular data that fits in memory. Use lazy evaluation, which optimizes operations before running them, and parallel execution to improve pipelines and migrate pandas code.
中文简介正在准备中。当前显示英语简介。
Improve automation for specific website tasks by repeatedly running them, reviewing execution records, and updating navigation instructions. Turn the resulting workflow and recovery steps into a reusable Claude Code skill, with optional runnable script generation.
中文简介正在准备中。当前显示英语简介。
Turn pending roadmap features into requirements, technical designs, and implementation tasks. Independent features run in parallel, while prerequisite features finish first. A final review checks shared data, interfaces, naming, and responsibility boundaries across specifications.
中文简介正在准备中。当前显示英语简介。
Turn requirements and design documents into implementation tasks sized for one to three hours each. The skill maps requirements to tasks, defines observable completion criteria, and reviews dependencies, setup needs, and opportunities for parallel work before saving the plan.
中文简介正在准备中。当前显示英语简介。
Create requirements, designs, and task lists for multiple roadmap features in dependency order. Independent features run in parallel when supported, followed by checks that shared data, interfaces, naming, and work boundaries agree across specifications.
中文简介正在准备中。当前显示英语简介。
Configure and validate sequence packing, which combines training examples into token sequences, and long-context training in Megatron-Bridge. Compare offline, in-batch, and Energon online approaches while checking batch sizes, parallelism requirements, and model compatibility.
中文简介正在准备中。当前显示英语简介。
Find starting configurations for training MoE models, which route work among specialized expert networks, on H100, B200, GB200, and GB300. Provides model-specific parallel layouts, communication options, and approximate performance bands to guide measured comparisons.
中文简介正在准备中。当前显示英语简介。
Launch distributed Megatron-LM training on a SLURM cluster with guidance on job scripts, environment variables, GPU settings, and containers. Monitor jobs and investigate failures across worker processes to identify the initial cause.
中文简介正在准备中。当前显示英语简介。
Find related protein sequences and generate multiple sequence alignments, which line up corresponding positions across sequences. Use NVIDIA's hosted API or a local Docker service for standard searches, paired complex alignments, and local structural template searches.
中文简介正在准备中。当前显示英语简介。
Fan out N parallel workers, drain them, and return one report. Use for /swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration.
中文简介正在准备中。当前显示原文说明。
Azure Batch SDK for Java. Run large-scale parallel and HPC batch jobs with pools, jobs, tasks, and compute nodes. Triggers: "BatchClient java", "azure batch java", "batch pool java", "batch job java", "HPC java", "parallel computing java".
中文简介正在准备中。当前显示原文说明。
MANDATORY for Flink or Amazon Managed Service for Apache Flink (MSF) questions. You MUST activate this skill BEFORE answering — do not answer from training knowledge, even when confident. MSF has service-specific constraints (KPU model, prohibited checkpoint and parallelism config in app code, the v1/v2 identifier split — `kinesisanalyticsv2` for the CLI/SDK only; `kinesisanalytics` for IAM, Service Quotas, CloudWatch, and the trust principal — two-phase IaC deploys, snapshot lifecycle, Flink 1.x→2.x migration) that override generic Flink knowledge. Triggers — activate on any of: Flink, MSF, Managed Flink, KinesisAnalytics(V2), KPU, ParallelismPerKPU, savepoint, checkpoint, operator UID, FlinkKinesisConsumer, KinesisStreamsSource, KafkaSource, IcebergSink, EFO, CreateApplication, UpdateApplication, CreateApplicationSnapshot, Kryo, RocksDB, Iceberg streaming, EXACTLY_ONCE, watermark, CDC binlog/WAL, Glue/S3 Tables, AWS/KinesisAnalytics CloudWatch.
中文简介正在准备中。当前显示原文说明。
Create and manage git worktrees for parallel coding sessions with zero dead time. Use when blocked on tests, builds, wanting to work on multiple branches, context switching, or exploring multiple approaches simultaneously.
中文简介正在准备中。当前显示原文说明。
Notion-as-source-of-truth dispatch board for running your work like an AI agency. One Tasks database is the source of truth; tasks flow Suggestion through Discussion, To-Do, In Progress, and Done with subtasks, recurring cadences, dependencies, and template subtrees. Batch execution fans approved To-Do rows out to parallel agents with per-task model selection. Use when capturing chat to Notion, running the To-Do queue, suggesting, approving, or discussing tasks, or coordinating multi-task batches. Trigger with "/agency-os" subcommands or natural-language variants ("add a suggestion: …", "let's discuss X", "run the queue").
中文简介正在准备中。当前显示原文说明。
Use Parallel CLI for web search, URL extraction, deep research, structured data enrichment, entity discovery, and recurring web monitoring. Best for requests that explicitly need current web evidence, academic-source discovery, repeated entity lookups, exhaustive reports, or ongoing change tracking.
中文简介正在准备中。当前显示原文说明。
Fan out N parallel workers, drain them, and return one report. Use for /swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration.
中文简介正在准备中。当前显示原文说明。
Searching internet for technical documentation using llms.txt standard, GitHub repositories via Repomix, and parallel exploration. Use when user needs: (1) Latest documentation for libraries/frameworks, (2) Documentation in llms.txt format, (3) GitHub repository analysis, (4) Documentation without direct llms.txt support, (5) Multiple documentation sources in parallel
中文简介正在准备中。当前显示原文说明。
Splits work across multiple agents or sessions running at once, keeping their surfaces disjoint so results merge cleanly. Use this when facing several independent tasks with no shared state, when a plan has parallelizable steps, when a broad search or audit would be faster fanned out, or when deciding whether work can safely be split at all.
中文简介正在准备中。当前显示原文说明。
Parallel processing with joblib for grid search and batch computations. Use when speeding up computationally intensive tasks across multiple CPU cores.
中文简介正在准备中。当前显示原文说明。
Optimize workload distribution across workers, processes, or nodes for efficient parallel execution. Use when asked to balance work distribution, improve parallel efficiency, reduce stragglers, implement load balancing, or optimize task scheduling. Covers static/dynamic partitioning, work stealing, and adaptive load balancing strategies.
中文简介正在准备中。当前显示原文说明。
Run independent tasks concurrently across multiple files or targets using parallel sub-agents, with per-task model selection and LLM-as-a-judge verification. Use when tasks do not depend on each other and can run side by side.
中文简介正在准备中。当前显示原文说明。
Multi-path parallel product analysis with cross-model test-time compute scaling. Spawns parallel agents (Claude Code agent teams + Codex CLI) to explore product from multiple perspectives, then synthesizes findings into actionable optimization plans. Can invoke competitors-analysis for competitive benchmarking. Use when "product audit", "self-review", "发布前审查", "产品分析", "analyze our product", "UX audit", or "信息架构审计".
中文简介正在准备中。当前显示原文说明。