metadata · datawithbaraa — Claude Code skill: what it does, how to install
This Claude Code skill builds or extends metadata for a data source. It runs when table and column details need to be gathered from source files and documentation.
MIT ★ 106 updated 2026-11-06
/plugin marketplace add DataWithBaraa/claude-code-data-engineering-skills
/plugin install data-engineering@data-with-baraaWhat metadata does
The skill reads a source’s data files, profile, example metadata, and documents selected by the data engineer. It writes a Python script to extract table and column names, types, and keys, then creates a metadata file that follows the example’s format.
Column descriptions come from the documents. When a document has no description for a column, the field stays empty and is listed in the report. The script reads columns as text first, so IDs keep their leading zeros. It also saves the extraction script and leaves the source data files unchanged.
What it helps with
- build metadata for every table in a source
- extract column names and types from data files
- populate column descriptions from documentation
- report columns with no documented description
Example prompts for the agent
- Build metadata for the
purchasingsource from its files, profile, and example. - Extend the source metadata using the documents I point you to.
- Create the metadata file and report any empty descriptions.
How to install metadata
1Plugin marketplace
/plugin marketplace add DataWithBaraa/claude-code-data-engineering-skills/plugin install data-engineering@data-with-baraaThis installs the whole data-engineering plugin with all its skills
2Skills CLI (Vercel)
npx skills add DataWithBaraa/claude-code-data-engineering-skills --skill metadata -a claude-code3Manually: put the files in place
~/.claude/skills/metadata/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/metadata/
1Skills CLI (Vercel)
npx skills add DataWithBaraa/claude-code-data-engineering-skills --skill metadata -a codex2Manually: put the files in place
.agents/skills/metadata/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add DataWithBaraa/claude-code-data-engineering-skills --skill metadata -a cursor2Manually: put the files in place
.cursor/skills/metadata/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install DataWithBaraa/claude-code-data-engineering-skills skills/metadata2Manually: put the files in place
.github/skills/metadata/SKILL.md1Skills CLI (Vercel)
npx skills add DataWithBaraa/claude-code-data-engineering-skills --skill metadata -a gemini-cli2Platform CLI
gemini skills install https://github.com/DataWithBaraa/claude-code-data-engineering-skills --path skills/metadataManually: put the files in place
.devin/skills/metadata/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about metadata
How do I install metadata in Claude Code?
Add the marketplace with /plugin marketplace add DataWithBaraa/claude-code-data-engineering-skills, then install the plugin with /plugin install data-engineering@data-with-baraa.
What does metadata do?
It extracts table names, column names, types, and keys from source files, then takes column descriptions from the selected documents. It creates a metadata file and reports descriptions that remain empty.
What does metadata need to work?
It needs the source data files, the source profile, an example metadata file, and documents selected by the data engineer. The skill also writes a Python script to extract technical details.
Source text
SKILL.md · MIT
--- name: metadata description: Use when the metadata of a source has to be built or extended, to describe every table and column, the names and types from the data with a script, the descriptions from the documents, never guessed. argument-hint: "[source]" --- # Metadata The job: extend the metadata of a source from its files and its documents. Facts from the data, descriptions from the documents, nothing invented. Input: the source, its data files, its documents, and one example table written by hand. ## Steps 1. Read the example: one source, one table, the format to follow. Table name, description, then every column with its name, type, key and description. 2. Point at the data of the source: the exports or the landing files, one file per table. 3. Write a Python script that reads every file, every column as text first, and extracts the table names, the column names and the data types. IDs stay strings and keep their leading zeros. The script does this, not the model, so nothing is invented. 4. Take the keys from the profile of the source, never from a guess. 5. Read only the documents the data engineer points at: PDFs, handbooks, data dictionaries. 6. Take every description from the documents. A description must trace to a document. 7. Where the documents say nothing, leave the description empty and list it. Never guess a description. 8. Write every table into the metadata file, in the format of the example. 9. Report what was filled and what is still empty. ## Output One metadata file per source, like [references/purchasing.yaml](references/purchasing.yaml): the source, its tables, and for every column the name, the type, the key and the description. ## Reads - the data files of the source, the profile, the documents, the example metadata ## Writes - the metadata file of the source, and the script that extracted the technical facts ## Never - Never invents a column, a type or a description. - Never reads a document the data engineer did not point at. - Never changes a data file.
© Data With Baraa · MIT · DataWithBaraa/claude-code-data-engineering-skills
Similar skills
All skills for Data and databases →metadata by other authors comes first.