OpenAI Changeset Validation
Overview
This skill validates whether changesets correctly reflect package changes and follow the repository rules. It relies on the shared prompt in references/validation-prompt.md so local Codex reviews and GitHub Actions share the same logic.
Experimental or preview-only feature additions that are explicitly labeled as such in the diff may remain a patch bump when they do not change existing behavior.
Major bumps are only allowed after the first major release; before that, do not use major bumps for feature-level changes.
Quick start
Local (Codex-driven):
- Run:
pnpm changeset:validate-prompt - Apply the rules from
references/validation-prompt.mdto the generated prompt. - Respond with a JSON verdict containing ok/errors/warnings/required_bump (English-only strings).
CI (Codex Action):
- Run:
pnpm changeset:validate-prompt -- --ci --output .github/codex/prompts/changeset-validation.generated.md - Use
openai/codex-actionwith the generated prompt and JSON schema to get a structured verdict.
Workflow
- Generate the prompt context via
pnpm changeset:validate-prompt. - Apply the rules in
references/validation-prompt.mdto judge correctness. - Provide a clear verdict and required bump (patch/minor/major/none).
- If the changeset needs edits, update it and re-run the validation.
Branch-level changeset rule (do not skip)
- If any
.changeset/*.mdfile is already present in the current branch diff, treat it as the active changeset. Do not add a new changeset file. - Instead, update the existing changeset file by rewriting its summary to reflect the primary change relative to
main. Do not try to capture every incremental update. - Keep the summary at one line; replace the line as needed rather than appending. If you must include multiple points, compress them into a single line.
Shared source of truth
- Keep the prompt file as the single source of validation rules.
- Keep the script lightweight: it should only gather context and emit the prompt.
Resources
references/validation-prompt.md