菜单
DeepEval pytest-style LLM testing patterns with built-in metrics, custom evaluators, and CI integration. Use when creating LLM tests, evaluating RAG quality, or measuring faithfulness/relevance.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | deepeval-testing |
| description | DeepEval pytest-style LLM testing patterns with built-in metrics, custom evaluators, and CI integration. Use when creating LLM tests, evaluating RAG quality, or measuring faithfulness/relevance. |
| allowed-tools | Bash, Read, Write, Edit, Grep, Glob, WebFetch |
Skill for pytest-style LLM evaluation with DeepEval.
DeepEval provides:
This skill is automatically invoked when:
| Script | Description |
|---|---|
scripts/init-deepeval.sh | Initialize DeepEval project |
scripts/run-tests.sh | Run DeepEval tests |
| Template | Description |
|---|---|
templates/conftest.py | pytest configuration |
templates/test_basic.py | Basic test structure |
templates/test_rag.py | RAG evaluation tests |
AnswerRelevancyMetricFaithfulnessMetricContextualRelevancyMetricHallucinationMetricToxicityMetricBiasMetricGoogle File Search API patterns for managed RAG with Gemini. Covers both TypeScript (@google/genai) and Python (google-genai) SDKs. Use when building File Search integrations, implementing RAG with Google AI, or when user mentions Google File Search, Gemini RAG, document indexing, or semantic search.
BullMQ queue configuration patterns including connection pooling, job options, rate limiting, and TypeScript types. Use when setting up queues or configuring job behavior.
BullMQ worker implementation patterns including job processors, concurrency, error handling, and graceful shutdown. Use when creating workers or handling job processing.
DigitalOcean Managed Database provisioning and connection patterns
DigitalOcean networking patterns - VPCs, firewalls, load balancers, DNS
DigitalOcean Spaces (S3-compatible) object storage patterns