Use this skill when the user wants to document a data pipeline. Triggers include: 'document this pipeline', 'write up what this ETL does', 'generate docs for this script', or when pipeline code or a description is provided and readable documentation is needed for handoff or onboarding.
How this skill is triggered — by the user, by Claude, or both
Slash command
/foundation-wide-skills:pipeline-documentationThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are helping a Gates Foundation data engineer or analyst document a data pipeline.
You are helping a Gates Foundation data engineer or analyst document a data pipeline.
Ask the user to provide one or more of:
Work with whatever is available; flag what is missing.
For each stage:
One paragraph readable by a non-technical stakeholder.
## Pipeline Documentation: [Name]
**Purpose:** [One paragraph overview]
**Owner:** [Team or person]
**Frequency:** [How often it runs]
### Data Sources
| Source | System | Format | Refresh cadence | Notes |
|---|---|---|---|---|
### Pipeline Stages
#### Stage 1 — [Name]
**What it does:** [One sentence]
**Input:** [Description]
**Transformations:**
- [Transform 1]
**Output:** [Description]
**Assumptions:** [What must be true]
### Output Schema
| Field | Type | Description | Nullable |
|---|---|---|---|
### Data Quality Notes
| Issue | Field(s) | Severity | Handling |
|---|---|---|---|
### Known Limitations
- [Limitation 1]
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub shchen-idmod/gf-agent-skills --plugin foundation-wide-skills