Skip to content

[feature-request] Banned-vocabulary detector for UI string literals #441

Description

@jessegall

Proposal:
The plain-names law is enforced on identifiers, but the same poetic vocabulary leaks into UI COPY — string literals passed to captions, headings, buttons and empty states — where no identifier sweep can see it. Example that survived every rename batch until the user caught it on screen: Section captions 'Taken'/'Given' over a trace step's payload, which are 'Input'/'Output'. A frontend/backend detector that checks string literals flowing into known text positions (SectionHeading, captionText, Button labels, Text::of in View components) against the project's banned-word list (said/asked/told/taken/given/whilst...) would catch these mechanically, the way the identifier detectors already do.

Filed via commandments feature-request from a consumer project.

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