| name | yaml-master |
| description | Execute proactive YAML intelligence: automatically activates when working with YAML files.
Use when appropriate context detected. Trigger with relevant phrases based on skill purpose.
|
| allowed-tools | Read, Write, Edit, Grep, Glob, Bash(general:*), Bash(util:*) |
| version | 2.21.0 |
| author | Jeremy Longshore <jeremy@intentsolutions.io> |
| license | MIT |
| tags | ["productivity","yaml-master"] |
| compatibility | Designed for Claude Code |
YAML Master
Proactive YAML intelligence: validate syntax, enforce consistent formatting, and keep configuration files schema-correct (Kubernetes, GitHub Actions, Docker Compose, and similar).
Overview
This skill activates when working with .yml/.yaml files to detect structural issues early (indentation, anchors, type mismatches), and to produce safe, minimal edits that keep CI/config tooling happy.
Prerequisites
- The YAML file(s) to inspect and their intended target (e.g., Kubernetes, GitHub Actions, Compose)
- Any relevant schema or constraints (when available)
- Permission to edit the file(s) (or to propose a patch)
Instructions
- Parse and validate YAML syntax (identify the first breaking error and its location).
- Normalize formatting (indentation, quoting) without changing semantics.
- Validate structure against the target system’s expectations (keys, types, required fields).
- Identify risky patterns (duplicate keys, ambiguous scalars, anchors used incorrectly).
- Output a minimal patch plus a short validation checklist (what to run next).
Output
- Corrected YAML with minimal diffs
- A concise list of issues found (syntax vs schema vs best practice)
- Follow-up validation commands appropriate for the target (e.g.,
kubectl apply --dry-run=client, CI lint)
Error Handling
- If the schema/target is unknown, ask for the target system and apply syntax-only fixes first.
- If the YAML is valid but tooling still fails, surface the exact downstream error and reconcile expectations.
Examples
Example: Fix an indentation/syntax error
- Input: a workflow with a mis-indented
steps: block.
- Output: corrected indentation and a note on which job/step was affected.
Example: Convert JSON to YAML safely
- Input: a JSON config blob.
- Output: YAML with explicit quoting where necessary to avoid type surprises.
Resources
