Harness for building data engineering agents designed around human control, correctness, and safety.
This repository also configures its own agents. See what sqrl tells them →
Skill Claude CodeCodex needs its repo
Use when configuring a DataSQRL project's -package.json files. Use for engine selection (Flink, Postgres, Kafka, Iceberg and its query engines DuckDB, Snowflake, Spark SQL, Redshift, Trino), API protocols, compiler options, runtime settings, or deployment/infrastructure settings.
Skill Claude CodeCodex
Use when a task provides source, sample, or test data files to inspect before modeling schemas, connectors, writing test data or implementing anything. Ensures you look at the real data (size, format, values) instead of guessing from filenames, READMEs, or prior knowledge.
Skill Claude CodeCodex
Use when designing or modifying the API exposed by a DataSQRL data pipeline. Use for GraphQL, REST, or MCP endpoints, access control, or data serving requirements. Use for adjusting exposed GraphQL schema.
Skill Claude CodeCodex
Use when the workspace contains a large data file, or when a compile/test run is blocked with LARGEDATABLOCKED. Used to sample the large files down or exclude them so that DataSQRL compiles seamlessly.
Skill Claude CodeCodex
Use when writing or editing DataSQRL scripts (.sqrl) for data processing, transformation, aggregation, or serving logic. Use for any pipeline logic implementation. Contains SQRL language spec.
Skill Claude CodeCodex
Use when SQRL/SQL alone cannot implement required logic. Use to create custom scalar, aggregate, or table functions in Java.
Skill Claude CodeCodex
Use when starting a new DataSQRL project or initializing an empty pipeline. Use for greenfield implementations, project scaffolding, or setting up the folder structure.
Skill Claude CodeCodex
Use when connecting to external data systems for ingestion or export in DataSQRL. Use for Kafka, filesystem, database, Iceberg, or other connector configuration.
Skill Claude CodeCodex
Use when adding or updating tests for a DataSQRL pipeline. Use after implementing features to verify correctness.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: