Chromestatus Adding A Field
Developmentchromestatus-adding-a-fieldChromestatus Adding A Field is an Agent Skill for development published in the GoogleChrome/chromium-dashboard repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
How to add a new field to a feature across the entire stack (Data, API, and Frontend).
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/GoogleChrome/chromium-dashboard/blob/HEAD/.agents/skills/chromestatus-adding-a-field/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/chromestatus-adding-a-field, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
Adding a Field to a Feature
This guide explains how to add a new field to a feature entry, ensuring it is properly persisted in the database, exposed via the API, and editable in the frontend.
1. Data Layer (Backend)
The primary data model for features is FeatureEntry located in internals/core_models.py.
Add Property to FeatureEntry
Add your new field as an ndb property.
# internals/core_models.py
class FeatureEntry(ndb.Model):
# ...
my_new_field = ndb.StringProperty() # Use appropriate ndb type (IntegerProperty, BooleanProperty, etc.)
2. API Layer (Backend)
Update api_specs.py
The FeaturesAPI uses api/api_specs.py to map request fields to database properties and handle types.
# api/api_specs.py
FEATURE_FIELD_DATA_TYPES: FIELD_INFO_DATA_TYPE = [
# ...
('my_new_field', 'str'), # 'str', 'int', 'bool', 'link', 'emails', etc.
]
Update Converters (Optional)
If your field needs custom formatting for the JSON response (e.g., date to string), update api/converters.py. Most simple types are handled automatically by to_dict.
# api/converters.py
def feature_entry_to_json_verbose(fe: FeatureEntry, ...):
# ...
d: VerboseFeatureDict = {
# ...
'my_new_field': fe.my_new_field,
}
3. OpenAPI Specification
Update openapi/api.yaml if you want the field to be part of the official API documentation and generated models.
# openapi/api.yaml
# (Find the relevant schema, e.g., Feature or FeatureEntry if exists,
# or update the endpoint response descriptions)
4. Frontend Layer
Define Form Field Metadata
Add the field's metadata (label, help text, component type) in client-src/elements/form-field-specs.ts.
// client-src/elements/form-field-specs.ts
export const ALL_FIELDS: Record<string, Field> = {
// ...
my_new_field: {
type: 'input', // 'textarea', 'checkbox', 'select', etc.
attrs: TEXT_FIELD_ATTRS,
label: 'My New Field',
usage: ALL_INTENT_USAGE_BY_FEATURE_TYPE,
help_text: html`Explain what this field is for.`,
},
};
Add to Form Definitions
Include the field in the relevant forms in client-src/elements/form-definition.ts.
// client-src/elements/form-definition.ts
export const FLAT_METADATA_FIELDS: MetadataFields = {
name: 'Feature metadata',
sections: [
{
name: 'Feature metadata',
fields: [
// ...
'my_new_field',
],
},
],
};
5. Verification
- Backend: Run
npm testto ensure NDB properties and converters work. - Frontend: Run
npm run webtestor manually verify the new field appears in the feature edit form. - API: Manually check
/api/v0/features/{id}to see if the field is present.