メニュー
Build and run evaluators for AI/LLM applications using Phoenix.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Build agentic applications with GitHub Copilot SDK. Use when embedding AI agents in apps, creating custom tools, implementing streaming responses, managing sessions, connecting to MCP servers, or creating custom agents. Triggers on Copilot SDK, GitHub SDK, agentic app, embed Copilot, programmable agent, MCP server, custom agent.
OpenInference semantic conventions and instrumentation for Phoenix AI observability. Use when implementing LLM tracing, creating custom spans, or deploying to production.
Scaffold a complete Power Apps Code App project with PAC CLI setup, SDK integration, and connector configuration
Provides exact patterns for diagnosing and fixing automatic batching regressions in React 18 class components. Use this skill whenever a class component has multiple setState calls in an async method, inside setTimeout, inside a Promise .then() or .catch(), or in a native event handler. Use it before writing any flushSync call - the decision tree here prevents unnecessary flushSync overuse. Also use this skill when fixing test failures caused by intermediate state assertions that break after React 18 upgrade.
Provides exact Enzyme → React Testing Library migration patterns for React 18 upgrades. Use this skill whenever Enzyme tests need to be rewritten - shallow, mount, wrapper.find(), wrapper.simulate(), wrapper.prop(), wrapper.state(), wrapper.instance(), Enzyme configure/Adapter calls, or any test file that imports from enzyme. This skill covers the full API mapping and the philosophy shift from implementation testing to behavior testing. Always read this skill before rewriting Enzyme tests - do not translate Enzyme APIs 1:1, that produces brittle RTL tests.
Provides the complete migration pattern for React legacy context API (contextTypes, childContextTypes, getChildContext) to the modern createContext API. Use this skill whenever migrating legacy context in class components - this is always a cross-file migration requiring the provider AND all consumers to be updated together. Use it before touching any contextTypes or childContextTypes code, because migrating only the provider without the consumers (or vice versa) will cause a runtime failure. Always read this skill before writing any context migration - the cross-file coordination steps here prevent the most common context migration bugs.
| name | phoenix-evals |
| description | Build and run evaluators for AI/LLM applications using Phoenix. |
| license | Apache-2.0 |
| compatibility | Requires Phoenix server. Python skills need phoenix and openai packages; TypeScript skills need @arizeai/phoenix-client. |
| metadata | {"author":"oss@arize.com","version":"1.0.0","languages":"Python, TypeScript"} |
Build evaluators for AI/LLM applications. Code first, LLM for nuance, validate against humans.
Starting Fresh: observe-tracing-setup → error-analysis → axial-coding → evaluators-overview
Building Evaluator: fundamentals → common-mistakes-python → evaluators-{code|llm}-{python|typescript} → validation-evaluators-{python|typescript}
RAG Systems: evaluators-rag → evaluators-code-_ (retrieval) → evaluators-llm-_ (faithfulness)
Production: production-overview → production-guardrails → production-continuous
| Prefix | Description |
|---|---|
fundamentals-* | Types, scores, anti-patterns |
observe-* | Tracing, sampling |
error-analysis-* | Finding failures |
axial-coding-* | Categorizing failures |
evaluators-* | Code, LLM, RAG evaluators |
experiments-* | Datasets, running experiments |
validation-* | Validating evaluator accuracy against human labels |
production-* | CI/CD, monitoring |
| Principle | Action |
|---|---|
| Error analysis first | Can't automate what you haven't observed |
| Custom > generic | Build from your failures |
| Code first | Deterministic before LLM |
| Validate judges | >80% TPR/TNR |
| Binary > Likert | Pass/fail, not 1-5 |
