Skip to content

Improve turbo:check granularity and content validation #21

@gtbuchanan

Description

@gtbuchanan

Summary

turbo:check currently only validates existence of turbo tasks and
package scripts. It doesn't check content or distinguish between task
and script ignores.

Current limitations

  • Existence-only: only flags missing tasks/scripts, not modified
    content. Can't distinguish intentional overrides from accidental
    deletions.
  • No granularity: --ignore compile:ts ignores both the turbo
    task and the package script, even if only the script is intentionally
    different.
  • CLI-only: ignores must be passed as args every time, not
    persisted in config.

Possible improvements

  • Validate task/script content against expected baseline
  • Separate --ignore-task and --ignore-script flags
  • Support a .turbocheck or turbo:check field in package.json for
    persisted ignores
  • Report modified values as info (not drift) so consumers see what
    they've customized

Context

Added during #17 (Turborepo integration). The current implementation
is sufficient for detecting missing config but not for auditing
intentional deviations.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions