agentleFS
TheophilusChinomonaTheophilusChinomona/idev/skills/feature-completeness/SKILL.md

feature-completeness

Skill0 starsChanged 5 months agoMITView on GitHub

Traces a feature end-to-end (UI, hook, service, API, controller, database) to verify every link is wired. Use after completing a multi-layer feature or when asked whether a feature is fully connected.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 8 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    Verify that a newly created feature is fully wired end-to-end — from UI component through hooks/services to API endpoint to backend …

    line 9
  2. Distinctive:

    After completing a multi-layer feature (FE + BE)

    line 12
  3. Distinctive:

    After creating a new page/container that calls an API

    line 13
  4. Distinctive:

    When user reports "it builds but doesn't work"

    line 14
  5. Distinctive:

    Do NOT run for single-file changes — only for multi-layer features

    line 225
  6. Distinctive:

    Do NOT trace into external libraries — stop at the project boundary

    line 226
  7. Distinctive:

    Do NOT check every CRUD operation individually — check the pattern once

    line 227
  8. Distinctive:

    Do NOT block on database table not existing — warn and continue

    line 228
  9. Distinctive:

    Do NOT re-read files already checked by other skills — use cached results

    line 229

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Feature Completeness Skill

Purpose

Verify that a newly created feature is fully wired end-to-end — from UI component through hooks/services to API endpoint to backend controller/service/database. Catches disconnected pieces that compile but don't actually work together.

Activation

  • After completing a multi-layer feature (FE + BE)
  • After creating a new page/container that calls an API
  • When user reports "it builds but doesn't work"

How It Works

This skill traces the full request path from UI to database and back, verifying each connection point. It's technology-agnostic — it detects the layers from the architecture-scanner cache.


Phase 1: Identify the Feature Chain

For a full-stack feature, the chain is:

Generic chain (adapts to any stack):

  UI Layer         → User sees/clicks something
    ↓
  State/Hook Layer → Manages data fetching/mutation
    ↓
  Service Layer    → Makes HTTP call to API
    ↓
  API Layer        → Receives request, routes to handler
    ↓
  Business Layer   → Processes logic, validates, transforms
    ↓
  Data Layer       → Reads/writes database
    ↓
  Database         → Table exists with correct schema

Stack-specific chains:

React + .NET:
  Component → useQuery/useMutation → service.ts → API URL →
  Controller.cs → Service.cs → Repository → DbContext → SQL Table

React + Express:
  Component → useQuery/useMutation → service.ts → API URL →
  router.ts → controller.ts → service.ts → Prisma/Sequelize → SQL Table

React + FastAPI:
  Component → useQuery/useMutation → service.ts → API URL →
  router.py → service.py → SQLAlchemy → SQL Table

Vue + Django:
  Component → composable → api.ts → API URL →
  views.py → serializer.py → model.py → SQL Table

Phase 2: Trace Each Link

For each link in the chain, verify it exists and connects correctly:

Link 1: UI → Hook/State

Check: Does the component import and call the hook?
  Grep component file for hook name (e.g., useGetOrders)
  Verify: hook is called, return value is used (data, isLoading, error)
  Fail if: hook is imported but never called, or return values are ignored

Link 2: Hook → Service

Check: Does the hook call the service function?
  Read hook file → find queryFn or mutationFn
  Verify: service function is imported and called with correct params
  Fail if: service function doesn't exist or signature doesn't match

Link 3: Service → API URL

Check: Does the service make an HTTP call to the correct URL?
  Read service file → find axios/fetch call
  Verify: URL matches a known backend route (check architecture-scanner cache)
  Fail if: URL has no matching backend route (a recurring bug class — see the
  lessons-learned skill's log for project-specific instances)

Link 4: API URL → Controller

Check: Does the backend controller have a matching route?
  Grep controllers for the URL path
  Verify: HTTP method matches (GET/POST/PUT/DELETE)
  Fail if: no controller route matches

Link 5: Controller → Service

Check: Does the controller call the service method?
  Read controller method → find service call
  Verify: service interface is injected and method exists
  Fail if: service method doesn't exist or parameters don't match

Link 6: Service → Repository/DB

Check: Does the service access the database?
  Read service method → find repository/DB calls
  Verify: entity type is correct, query filters are present
  Fail if: repository for this entity isn't registered (a recurring bug class —
  see the lessons-learned skill's log for project-specific instances)

Link 7: Database → Table

Check: Does the database table exist?
  Check DbContext for DbSet declaration
  Check for database change script in .claude/idev/database-changes/
  Warn if: DbSet exists but no table creation script found

Phase 3: Report Results

Feature Completeness: Orders CRUD

  Chain: UI → Hook → Service → API → Controller → Service → DB

  ✓ Link 1: OrdersTabContainer → useGetOrders (called, data used)
  ✓ Link 2: useGetOrders → ordersService.getAll (imported, called)
  ✓ Link 3: getAll → GET api/orders/get-all (URL found in architecture cache)
  ✓ Link 4: api/orders/get-all → OrdersController.GetAll (route match)
  ✓ Link 5: GetAll → _orderService.GetAll (injected, method exists)
  ✓ Link 6: GetAll → _repository.GetAllAsync (IGenericRepository<Order> registered)
  ⚠ Link 7: DbSet<Order> exists, DB table pending DBA execution

  Result: Feature is wired end-to-end. Blocked on: database table creation.

Phase 4: Common Disconnection Points

Issues this skill catches that compile but fail at runtime:

1. Component renders but shows no data
   → Hook is called but service function returns wrong URL (404)

2. Service call succeeds but data shape is wrong
   → FE type doesn't match BE DTO fields (api-contract-validation catches this)

3. Controller receives request but service throws
   → Service depends on unregistered repository (post-creation-verify catches this)

4. Everything works but wrong data appears
   → Query filter missing the tenant filter column (multi-tenancy leak; e.g., a ClientId column)
   → Query missing the soft-delete flag check (e.g., a RecordStatusId column)

5. Feature works for create but not update/delete
   → Only some CRUD operations were wired
   → Missing mutation hooks or service functions

6. Page loads but route doesn't work
   → Component created but route not registered in the project's route registration file (e.g., Routes.tsx)

Phase 5: Completeness Checklist

Auto-generated based on detected stack:

Full-Stack CRUD Feature

□ FE Types: Request/Response interfaces defined
□ FE Service: All CRUD functions (getAll, getById, create, update, delete)
□ FE Hooks: useQuery for reads, useMutation for writes
□ FE Container: Imports hooks, handles loading/error/empty states
□ FE Route: Page registered in router (if new page)
□ BE Entity: Created with all required fields
□ BE DTOs: Return, Create, Update DTOs + mapping extensions
□ BE Interface: Service interface with all methods
□ BE Service: Implementation with error handling
□ BE Controller: All route handlers
□ BE DI: Service + Repository registered
□ BE DbContext: DbSet declared
□ DB: Table creation script exists
□ Contract: FE types match BE DTOs
□ Chain: All 7 links verified connected

Phase 6: Integration

Complete feature creation workflow:
  1. lessons-learned → prevention steps
  2. Create BE files (entity, DTOs, service, controller)
  3. post-creation-verify → registrations
  4. build-check → BE compiles
  5. Create FE files (types, service, hooks, container)
  6. build-check → FE compiles
  7. api-contract-validation → FE↔BE types match
  8. feature-completeness → end-to-end chain verified  ← THIS SKILL
  9. self-review → code quality
  10. Report to user

Anti-Patterns

  1. Do NOT run for single-file changes — only for multi-layer features
  2. Do NOT trace into external libraries — stop at the project boundary
  3. Do NOT check every CRUD operation individually — check the pattern once
  4. Do NOT block on database table not existing — warn and continue
  5. Do NOT re-read files already checked by other skills — use cached results

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.