Datus-ai

Operate a remote Apache Airflow 2.x or 3.x deployment (list/trigger/monitor DAGs, tasks, variables, connections, pools, backfills) and deploy DAG files to S3 or a dags folder via the `datus airflow` CLI

Configure an environment profile for the `datus airflow` plugin (API endpoint, credentials, optional S3/dags-folder deployment target)

Configure or troubleshoot an Alibaba Cloud ACK profile and its paired datus k8s provider profile, including the default credential chain, access keys or STS, RAM roles, RRSA/external credentials, endpoint selection, and verification.

Inspect Alibaba Cloud ACK clusters, node pools, add-ons, tasks, and authentication, and hand temporary ACK credentials to datus k8s. Use for ACK inventory, health, change diagnostics, add-on status, or Kubernetes access through an ACK provider profile.

Configure an environment profile for the `datus s3` plugin (AWS region + credentials, optional default bucket and SSE-KMS key)

Configure or troubleshoot an Azure Blob or ADLS Gen2 profile for datus adls, including Entra DefaultAzureCredential, service principals, managed identity, account key or SAS, sovereign clouds, HNS behavior, permissions, and verification.

Browse, inspect, preview, upload, download, copy, sync, move, delete, and sign Azure Blob or ADLS Gen2 paths, and inspect filesystems or manage ACLs with datus adls. Use for abfss:// data movement, artifacts, SAS URLs, or hierarchical namespace ACL operations.

Configure or troubleshoot an Azure Kubernetes Service profile and its paired datus k8s provider profile, including Azure cloud, DefaultAzureCredential, service principals, managed identity, private endpoints, and verification.

Lightweight project initialization — optionally scoped to specific files / tables / datasources / domains. Reads the project's SQL (through the persisted table lineage graph, upsert_lineage / query_lineage), human-written docs and the database schema without…

Dosi native OSI dataset, relationship, and metric authoring guidance

Build the project's vector-indexed knowledge base from files plus database metadata — optionally scoped to specific files / tables / datasources / domains. Scan the in-scope material, classify it into business domains, explore each domain's tables and docs in…

Mine shortest atomic facts from (question + gold_sql) pairs into ./knowledge/*.md; either by simulating SQL drafting (lite) or by driving the gen_sql subagent in blind iteration (deep)

Audit and reorganize every persistent store — semantic_models, metrics, reference_sql, knowledge, memory, AGENTS.md, skills — verifying each item sits in the correct store per storage-classify, and surfacing duplicates, misclassifications, conflicts, and…

Review the current chat session and persist its valuable takeaways — business facts/rules, validated SQL, metric/model definitions, durable preferences, project conventions, reusable workflows — by classifying each via storage-classify and routing it to the…

Decide where a produced artifact must be persisted before writing it, then route semantic_models and metrics through semantic_modeling, reference_sql through gen_sql_summary, knowledge through extract-knowledge (lite), memory through add_memory, skills…

Create or update custom Datus subagents by editing the loaded agent.yml. Use when a workflow needs persistent agentic_nodes scoped to already-built tables, metrics, or reference SQL.

Generate, review, and understand StarRocks SQL. Use for StarRocks queries, OLAP table DDL, DML, materialized views, catalogs, S3/GCS imports with FILES() or Broker Load, Stream Load, Routine Load, and rewrites where MySQL compatibility, table types,…

Generate, review, and understand Huawei Cloud GaussDB(DWS) SQL. Use for DWS queries, DDL, DML, table design, and SQL rewrites where PostgreSQL 9.2 compatibility, ORA/TD/MySQL compatibility-mode semantics, row/column storage, data distribution, partitioning,…

Generate or review Google BigQuery Standard SQL, especially when project.dataset qualification, nested data, time semantics, partition pruning, or BigQuery-specific DDL affect correctness or cost.

Generate and review Alibaba Cloud MaxCompute or ODPS SQL. Use when namespace mode, partition scanning, data type editions, write semantics, or transactional-table restrictions affect SQL correctness or cost.

Generate, review, and understand TiDB SQL. Use for TiDB queries and DDL where MySQL compatibility, unsupported constructs, silently-ignored clauses, or write-hotspot avoidance can affect correctness or performance.

Generate, review, and understand Apache Doris SQL. Use when the target engine is Apache Doris (not StarRocks or MySQL) for queries, OLAP DDL with the Duplicate/Unique/Aggregate key models, bucketing and partitioning, synchronous and asynchronous materialized…

Generate, review, and understand GaussDB / openGauss SQL. Use for GaussDB queries, DDL, DML, table design, and SQL rewrites where PostgreSQL compatibility, A/B/PG database compatibility modes, distributed deployments, or version-specific syntax matter.

Generate, review, and understand Alibaba Cloud Hologres SQL. Use for Hologres queries, DDL, DML, table design, foreign tables, data movement, and SQL rewrites where PostgreSQL compatibility, namespaces, storage properties, constraints, transactions, or…

Research, write, wire in, preview and PR one SEO blog post for datus.ai (the Datus-website repo) end to end, then record it so no topic is ever written twice. Use when the operator asks for a blog post — 写一篇 blog、从 /glossary 挑一个方向写文章、写一篇 SEO 博客、新增 blog…

Produce the weekly datus.ai SEO report from Google Search Console + GA4, cross-checked against the blog posts shipped that week, and hand the search opportunities to /datus-blog as topic input. Use when the operator asks for the SEO 周报, GSC / GA4 weekly…