Menu
Describe what the skill does and when it should trigger.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | your-skill-name |
| description | Describe what the skill does and when it should trigger. |
| version | 0.1.0 |
| metadata | {"internal":true} |
Read the required reference files first.
IntentAssumptionsResultValidationKaiwuDB 数据库源码编译和构建技能。用于从源码构建、编译或测试 KaiwuDB (KWDB)。使用 CMake 构建系统处理构建配置、单元测试、代码检查和安装。触发条件：(1) 编译询问："编译 KaiwuDB"、"构建 kwdbts2"、"cmake 构建"、"清理构建"；(2) C++ 单元测试询问："C++ 单元测试"、"运行C++单元测试"、"run C++ unittest"；(3) Go 单元测试询问："golang 单元测试"、"Go 单元测试"、"run golang test"。重要：执行任何操作前必须向用户确认所有配置选项，不得跳过询问步骤。
Automates end-to-end anomaly detection for time-series data stored in KaiwuDB / KWDB. Use this skill whenever the user mentions: - anomaly detection, outliers, or unusual patterns in KWDB / KaiwuDB time-series data - inspecting sensor metrics, IoT telemetry, or monitoring data for spikes, dips, or drift - "find anomalies", "detect outliers", "3-sigma check", "STL decomposition", or "time-series anomaly" - analyzing historical trends, abnormal points, or data quality issues in TS tables Even if the user does not explicitly say "anomaly", trigger this skill when they ask to inspect, validate, or flag unusual values in time-series columns (integer, float, double).
Optimize SQL query performance for KaiwuDB time-series and relational engines. Covers: EXPLAIN analysis, time-series optimization, pagination, cross-model queries. Trigger keywords: optimize query, slow query, explain, execution plan, performance, 性能, 查询优化. NOT for: DDL, schema design, deployment, DML writes.
Use when diagnosing KWDB incidents from logs, metrics, or system evidence, especially crashes, OOM, slow SQL, restarts, and cluster-wide availability symptoms.
Run KaiwuDB inspection and health-check tasks. Use this skill for database health checks, metrics collection, anomaly detection, and inspection report generation.
Design KWDB schemas and minimal DDL for relational, time-series, and mixed workloads.