Instructions file CodexOpenCode
Project guidance for WackyGem/Fleur, a multi-part software project with Python data tools, Rust engines, a frontend, deployment files, and documentation. It describes the project layout and where key technical information is kept.
Skill Claude CodeCodex
A data-dictionary workflow for maintaining field definitions and schemas across the Fleur data pipeline. A schema describes the fields, types, order, and missing-value rules that data follows between systems.
Skill Claude CodeCodex
A runbook for replaying past data through Dagster pipelines, which are scheduled workflows for processing data. It describes how to use the project’s `dg launch` commands and shared backfill jobs.
Skill Claude CodeCodex
A guide for turning one remote HTTP link into a data pipeline that stores Parquet files in S3 and raw data in ClickHouse. Parquet is a column-based file format, while S3 is object storage.
Skill Claude CodeCodex
A preparation guide for writing dbt staging models, which are cleaned and organized data tables built from raw inputs. It requires profiling the raw data before deciding on cleaning rules and tests.
Skill Claude CodeCodex
A maintenance guide for the Fleur codebase, covering its architecture, documentation, tests, and long-running data work.
Skill Claude CodeCodex
Use this skill for exploratory frontend debugging in fleur through an existing Chromium browser exposed by vnc-mini-desktop over Chrome DevTools Protocol.
Skill Claude CodeCodex
A procedure for keeping Fleur’s version information accurate and tracing releases back to their source files and commits. A commit is a saved snapshot in Git, the version-control system.
Skill Claude CodeCodex
A workflow guide for using separate Git worktrees, which are extra checkouts of one repository, when several coding tasks run at the same time.