搜索

PythonReactSEOPDF

文档写作 AI 技能

发现 文档写作 相关 Skills,比较来源仓库和用途,查看适合你工作流的技能说明。

分类目录: 文档写作

显示第 68,593–68,616 个,共 83,400 个

文档写作 清除筛选

全部技能

GitHub

train-debug

rHedBull

Interactive diagnostic workflow for training problems. Use when training is failing, loss is stuck, gradients explode, NaN occurs, or convergence is poor.

GitHub

train

rHedBull

Execute a neural network training run with mandatory monitoring and best-practice defaults. Use when user wants to train a model, start training, or run a training job.

GitHub

experiment-plan

rHedBull

Design rigorous experiments before running them. Use when planning experiments, testing hypotheses, or designing ablation studies.

GitHub

monitor

rHedBull

View training progress in real-time. Use when checking on running training, viewing live metrics, or opening dashboards.

GitHub

run-validation

rHedBull

Run model validation on a dataset. Use when testing model performance, comparing checkpoints, or running final evaluation.

GitHub

interpret-curves

rHedBull

Analyze training logs and explain what's happening. Use when reviewing training runs, understanding loss curves, or diagnosing completed training.

GitHub

scaling-analysis

rHedBull

Run scaling experiments to understand model/data/compute relationships. Use when investigating scaling laws, compute-optimal training, or model size decisions.

GitHub

typescript

Pallepadehat

TypeScript strict patterns and best practices. Trigger: When implementing or refactoring TypeScript in .ts/.tsx (types, interfaces, generics, const maps, type guards, removing any, tightening unknown).

GitHub

ai-sdk-5

Pallepadehat

Vercel AI SDK 5 patterns. Trigger: When building AI features with AI SDK v5 (chat, streaming, tools/function calling, UIMessage parts), including migration from v4.

GitHub

claude-agent-sdk

tkwn2080

Build production AI agents with the Claude Agent SDK. Use when working with: (1) Creating autonomous agents with query() or ClaudeSDKClient, (2) Defining custom tools via in-process MCP servers, (3) Implementin...

GitHub

gtm-ops

accoladians

Trigger when user mentions GTM, Tag Manager, Enhanced Conversions, or publishing. Guide through GTM operations.

GitHub

pmax-optimization

accoladians

Activate when user asks about Performance Max campaigns, search themes, asset groups, or PMax optimization. Provides guidance on managing PMax campaigns via xwander-ads plugin.

GitHub

gaql-queries

accoladians

Activate when user needs custom Google Ads data analysis, advanced reporting, or specific metric queries. Provides GAQL query building and execution guidance.

GitHub

conversion-tracking

accoladians

Activate when user asks about conversion tracking, HubSpot sync, offline conversions, or attribution. Provides guidance on conversion setup and troubleshooting.

GitHub

obsidian-todorust

mimiqdev

Todoist data retrieval for Obsidian using todorust CLI. Fetch tasks and projects in ready-to-use Markdown formats. Use when skills need to: (1) Get completed tasks for daily/weekly reports, (2) Query by project...

GitHub

ontology-phase-1-ingest

a4b-corporation

Phase 1 of Ontology Builder Pipeline. Ingests and catalogs all input materials from _input/ folder. Use when starting ontology building process or when processing new input documents for domain analysis.

GitHub

ontology-phase-4-generate

a4b-corporation

Phase 4 of Ontology Builder Pipeline. Generates final Ontology artifacts (entity definitions, workflow catalog, concept guides) from DRD. Use after Phase 3 DRD is complete.

GitHub

ontology-phase-2-analyze

a4b-corporation

Phase 2 of Ontology Builder Pipeline. AI acts as domain SME to analyze raw inputs, extract entities/workflows/rules, fill knowledge gaps using market expertise. Use after Phase 1 ingestion is complete.

GitHub

ontology-phase-3-synthesize

a4b-corporation

Phase 3 of Ontology Builder Pipeline. Synthesizes analysis results into complete Domain Requirement Document (DRD). Use after Phase 2 analysis is complete to generate structured DRD.

默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库,同一仓库的多个技能可能共享相同数值;它反映仓库关注度,不代表单个技能的安装量或质量评分。

更多筛选

热度筛选

热门标签