agentleFS
aiqadamaiqadam/qadam-flow/.agents/skills/add-feature/SKILL.md

add-feature

Skill5 starsChanged 39 days agoMITView on GitHub

Use when the user asks to add a feature, implement functionality, or build something spanning database, API, and frontend. ALWAYS use for multi-layer feature work.

28 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 1 is distinctive, and 11 are very specific to this project.

  • 0

    Common

    Most of the 26 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 11

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run the ubiquitous-language skill's overlap detection against .agents/features/ before proposing anything.

    line 12
  2. Very specific:

    No edition gating, ever: All features are available to all users.

    line 26
  3. Distinctive:

    Add a changeset (patch for a fix, minor for a new export)

    line 33
  4. Very specific:

    Read .agents/features/<module-name>.md first (e.g.

    line 37
  5. Very specific:

    Entity: follow the add-entity skill.

    line 39
  6. Very specific:

    Register entity: add to getEntities() in database-connection.ts (REQUIRED — TypeORM doesn't auto-discover)

    line 40
  7. Very specific:

    Migration: follow the db-migration skill — it is mandatory, not a reference.

    line 41
  8. Very specific:

    Controller: follow the add-endpoint skill.

    line 43
  9. Very specific:

    Project ownership: Add entitiesMustBeOwnedByCurrentProject hook if returning project-scoped data

    line 44
  10. Very specific:

    Add to SystemJobName or WorkerJobType enum in shared

    line 50
  11. Very specific:

    Add handler, register in app.ts via systemJobHandlers.registerJobHandler()

    line 51
  12. Very specific:

    Read the design skill first — any new surface follows it

    line 55

“Common” is measured: kept by at least 6 in 10 of the 26 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Add Feature End-to-End

Implement feature described in $ARGUMENTS across the full stack.

Step 0: Check it does not already exist

Run the ubiquitous-language skill's overlap detection against .agents/features/ before proposing anything. A feature that duplicates an existing one is the most expensive thing you can ship here, and the registry exists to catch it in a minute.

Then answer, before writing code:

  1. Does a narrower skill own a step of this? add-entity, add-endpoint and db-migration are the authority on their own steps — this skill sequences them, it does not replace them. Read each one when you reach its step.
  2. Project-scoped or platform-scoped? → every query filters by projectId or platformId accordingly (.agents/rules/data-isolation.md).
  3. Need a Permission? → add to the Permission enum in packages/shared/src/lib/core/common/security/permission.ts.
  4. Must work embedded? → check EmbeddingState in the frontend.
  5. No edition gating, ever. All features are available to all users; there is no EE tree, no paywall and no plan flag to hide behind (.agents/rules/edition-safety.md).

Step 1: Shared Types (packages/shared)

  • Define Zod schemas + z.infer types in src/lib/{domain}/
  • Export from src/index.ts barrel
  • Add a changeset (patch for a fix, minor for a new export)

Step 2: Server (packages/server/api)

Read .agents/features/<module-name>.md first (e.g. .agents/features/tables.md for the tables module).

  • Entity: follow the add-entity skill. Short version: EntitySchema + BaseColumnSchemaPart + ApIdSchema, see tables/table/table.entity.ts.
  • Register entity: add to getEntities() in database-connection.ts (REQUIRED — TypeORM doesn't auto-discover)
  • Migration: follow the db-migration skill — it is mandatory, not a reference.
  • Service: Factory (log: FastifyBaseLogger) => ({...}) if logging needed, plain object otherwise. See tables/table/table.service.ts.
  • Controller: follow the add-endpoint skill. FastifyPluginAsyncZod, route configs AFTER the controller, securityAccess required. See tables/table/table.controller.ts.
  • Project ownership: Add entitiesMustBeOwnedByCurrentProject hook if returning project-scoped data
  • Module: register it in app.ts alongside the existing modules — there is no CE/EE split to place it in
  • Side effects: Separate *-side-effects.ts file if mutations trigger events/webhooks

Step 3: Worker (if queued work needed)

  • Add to SystemJobName or WorkerJobType enum in shared
  • Add handler, register in app.ts via systemJobHandlers.registerJobHandler()

Step 4: Frontend (packages/web)

  • Read the design skill first — any new surface follows it
  • Feature folder: src/features/{feature}/api/, hooks/, components/
  • API client: See features/tables/api/tables-api.ts
  • Hooks: See features/tables/hooks/table-hooks.ts
  • Route: React.lazy() + ProjectRouterWrapper() + RoutePermissionGuard + SuspenseWrapper
  • Translations: en plus ru/uz/kk in the same PR (npm run check-i18n enforces parity)
  • Feature flags: flagsHooks.useFlag() or <FlagGuard>

Step 5: Tests

  • API test: packages/server/api/test/integration/ce/{feature}.test.ts
  • Use setupTestEnvironment() + createTestContext(app)

Step 6: Verify, then get it reviewed

npm run lint-dev
npm run typecheck     # required once you touched packages/web — vite build does not type-check
npm run check-i18n    # required once you touched translations
npm run test-api

Then delegate code-quality and app-sec per .agents/rules/agent-delegation.md, before you report the feature done. A multi-layer feature is exactly the shape of change both reviewers exist for.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.