데이터·AI 스킬

写中文的时候直接绕开 AI 腔，而不是写完再洗。给 LLM 当写作指令用：喂进题目和文体， 产出的初稿就已经没有模板句式、没有价值拔高骨架、没有等厚段落。 这是六段流水线的第一段。第一段决定文章有没有人味，后面几段只能修，补不出来。 这份文件本身刻意写成它要求的那种中文 —— 指令文档的版式和语气会被模型当范文模仿， 一份满是加粗和编号清单的规范，教出来的就是加粗和编号清单。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use when building, migrating, or debugging Agent Evals on Inngest: scoring AI agent or workflow outcomes, deferred scorers, sessions, traces, step experiments, experiment variant attribution, Insights queries, or production eval loops for prompts, models, tools, providers, and agent behavior. Covers TypeScript SDK v4 scoring beta APIs, `scoreMiddleware`, `step.score`, `inngest.score`, `createScorer`, `defer`, `group.experiment`, `experimentRef`, `meta.sessions`, and when to use durable workflow primitives for outcome-based evaluation.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use when streaming durable workflow updates to a UI in real time — live order status pages that animate as steps complete, AI agent token streaming from a function to the browser, log tailing for long-running jobs, or human-in-the-loop approval flows that publish a prompt and wait for a user reply. Covers Inngest v4 native realtime: defining typed channels, publishing from inside step.run, minting subscription tokens via server actions, and consuming the stream from React/Next.js client components.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use when adding cross-cutting concerns to durable functions — structured logging or tracing across all functions, error tracking with Sentry, payload encryption for sensitive data, dependency injection of clients (DB, Stripe, etc.) into function handlers, custom telemetry, or behavior that should apply uniformly across many functions. Covers Inngest middleware lifecycle, creating custom middleware, dependencyInjectionMiddleware, @inngest/middleware-encryption, @inngest/middleware-sentry, and custom middleware patterns.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

VictoriaMetrics provider for obz. Covers all 6 metric commands (query, list, info, labels, label-values, series) using MetricsQL, a PromQL superset with extensions like keep_metric_names and range_normalize. This skill should be used when the user mentions "VictoriaMetrics", "MetricsQL", "obz metric -p vm", or needs to query metrics from a VM backend.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

VictoriaLogs provider for obz. Supports log search using LogsQL, a purpose-built query language with _msg, _time, and _stream filters plus pipes for transformation. This skill should be used when the user mentions "VictoriaLogs", "LogsQL", "obz log search -p vl", or needs to search logs from a VL backend.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Query Strava fitness data including activities, athlete stats, segments, routes, clubs, and gear. Use when the user asks about cycling, running, swimming, workouts, training, or Strava data.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

DNS C2 operator skill for the AgentCore sandbox breakout demo. This skill should be used when the user asks to run C2 verification, attack a victim chatbot, send commands to a compromised session, retrieve exfiltrated data, or interact with the DNS C2 system.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Evaluate code (especially LLM-generated) for structural simplicity using Rich Hickey's "Simple Made Easy" framework. Use this skill whenever reviewing a PR, diff, or code snippet for accidental complexity — particularly when the code was generated by an AI coding assistant and line-by-line review isn't feasible. Also use when the user asks about complecting, simplicity vs. easiness, structural coupling, or concept deduplication. Trigger on phrases like "is this simple", "does this complect", "review for complexity", "structural analysis", or any reference to Hickey, Simple Made Easy, or grey-box review.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Use this when setting up a GitHub repo to push its Nix builds to Juspay's shared OSS Attic cache (cache.nixos.asia/oss) — adds the substituter to flake.nix, a nix-cache.yml GitHub Actions workflow, and prompts for the ATTIC_TOKEN secret.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

fMRI data analysis — preprocessing, GLM modeling, and group-level inference for task and resting-state studies.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Fine-tune large language models — data curation, SFT, parameter-efficient methods (LoRA), evaluation, and avoiding common training failures. Use when a base model needs task-specific behavior or domain knowledge.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Work with text embeddings — choosing models, pooling, similarity measures, evaluation, and multilingual considerations. Use when building semantic search, clustering, or any similarity-based feature.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Build evaluation frameworks for LLM systems — harness design, graders, datasets, regression tracking, and human-in-the-loop eval. Use when you need systematic measurement of model or agent quality.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Statistical analysis of extreme weather — EVT, return periods, thresholds, and detection/attribution of rare events.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Writing figure captions that make figures self-contained — what to include, ordering, and statistical reporting.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Programmatic form completion — web forms, PDF AcroForms, and data mapping — use when automating repetitive form filling.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Drizzle ORM guidance — schema-as-code, relational queries, migrations, and SQL-first TypeScript data access.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Build rich visualizations with Apache ECharts: option config, dataset transforms, large-data performance, and custom rendering. Use for complex dashboards and big datasets.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Building robust document ingestion — chunking, metadata, and multi-format pipelines for search and RAG.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Electronic health record analysis — phenotyping, missingness, bias, and observational study design with EHR data.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Distill large models into smaller ones — teacher-student training, data generation, loss design, and quality validation. Use when you need big-model behavior at small-model cost.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Generate and edit images with diffusion models: prompting, seeds, CFG, img2img, inpainting, conditioning, and API integration. Use when adding AI image generation to apps or workflows.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Compare files, text, and data effectively with diff strategies, semantic comparison, and review workflows.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.