菜单
Ensures test coverage metrics, logging context formats (text/json), exception safety, and performance constraints.
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 | quality-expert |
| description | Ensures test coverage metrics, logging context formats (text/json), exception safety, and performance constraints. |
This skill ensures exceptional code reliability, rigorous logging formats, performance benchmarking, and strict exception boundaries.
text formats and machine-parsable json structures depending on the environmental APP_LOGS_LEVEL settings.AppConfig.set_logger(). Include necessary context keys (model, output_format, page):logger = app_conf.set_logger("ui", default_context={"page": "cards"})
try:
rag.read_document()
except Exception as e:
st.error(f"Erreur lors de l'indexation : {e}")
search_on_mtd and search_on_ressources must have configurable execution timeouts.k=4) to prevent context window bloat and latency spikes.Manages continuous integration configurations, Ruff lint/formatting runs, local automated Pytest suites, and dependency alignments.
Checks medical data validity, oncology guidelines compliance (TNCD), and verifies patient record grounding to ensure clinical safety.
Manages local docker compose services (Milvus, Mongo), hardware VRAM allocations, GPU configurations, and network isolation.
Designs system prompts, shapes specialized agent personas, configures output schemas, and sets up LLM reasoning and retry behaviors.
Aligns the technical roadmap, patient dashboard interface, and case prioritization logic with French oncology clinical standards.
Develops, modifies, and debugs Python services, LangChain chains, Vector Store connectors, and Streamlit dashboards within the Oncoflow application.