Spec-to-Test Feature Implementation
Turn feature requests into approved specs, traceable tests, and scoped implementation with explicit completion evidence.
Before & after
Same prompt, same agent — the only difference is the skill.
That's a good feature idea. You'll want to think about what resetting means—does the user confirm first? What data gets reset? How long does the operation take? What happens if it fails? I could help you work through the requirements. There are probably some edge cases to consider too, like what if a user's default preferences haven't been set yet. I'd suggest documenting the happy path and then thinking about error scenarios. Would you like me to outline some acceptance criteria?
FEATURE SPECIFICATION: PREFERENCES RESET REQUIREMENT Users can restore all account settings to system defaults with a single action, reverting customizations made over time without affecting data or content. ACCEPTANCE CRITERIA • User can trigger reset from Settings > Advanced > Reset all preferences • Confirmation dialog requires user to type "RESET" to proceed • Operation completes within 2 seconds • POST /api/v1/users/me/preferences/reset requires authenticated session • Returns HTTP 200 with diff object showing what changed • All 47 preference keys revert to system defaults (documented in DEFAULTS.json) • User receives email confirmation within 30 seconds • Event "preferences_reset" logged with user_id, timestamp, ip_address EDGE CASES & SOLUTIONS • User clicks reset twice: second call returns 409 "already in progress" (prevents race) • Partial failure (email fails but reset succeeds): Return 202 partial success, queue retry • User has no reset permission: Return 403 with reason (e.g., organization-managed account) • Old defaults don't exist: Use hardcoded FALLBACK_DEFAULTS, log as "system_default_missing" ROLLBACK PLAN • Preference history maintained in audit_log for 90 days • User can restore any previous state via Settings > Preference history • Undo available for 24 hours after reset
About this skill
“The tests pass, but that isn't what I asked for.” This workflow turns a feature request into an agreed behavioral specification, derives tests from that specification, and keeps implementation tied to the approved scope. Ambiguities become visible decisions instead of hidden assumptions in code.
What it does
- Converts vague requests into observable behavior, explicit boundaries, and testable acceptance criteria.
- Finds contradictions in permissions, state transitions, failure handling, retries, and boundary conditions before implementation.
- Maps stable acceptance-criterion IDs to concrete test inputs and expected outcomes.
- Uses a specification approval checkpoint before generating executable tests or implementation code.
- Implements small behavioral slices and distinguishes requirement changes from implementation fixes.
- Reports which requirements passed, failed, remain untested, or were explicitly deferred.
What you get
A reviewable feature specification with acceptance criteria and a test map. When implementation is authorized, you also get tests, scoped code changes, and a completion report tied to the approved requirements, with execution evidence clearly separated from planned checks.
Best for
Developers and teams building features where ambiguous behavior, changing requirements, or tests that merely mirror the implementation keep causing rework.
SKILL.md preview
---
name: spec-to-test-feature-implementation
description: Turn feature requests into approved behavioral specifications, derive traceable tests, and implement against that baseline with explicit change control and verification evidence. Use when the user says "spec this before coding", "define acceptance criteria", "generate tests from this spec", "write the tests first", "implement the approved spec", "my tests pass but the feature is wrong", "requirements keep changing halfway through", or "check whether this implementation matches the spec".
---
# Spec-to-Test Feature Implementation
… (sign up to view the full skill)More development skills
View all Development skills →Performance Bottleneck Profiling: Node, Python, Go
Classifies a slowdown, gives exact profiler commands, reads the output, and proves the fix with same-conditions p95/p99 numbers.
Feature Scaffolding with Working Tests
Turn a short feature spec into code, runnable tests, and integration wiring that match your existing project.
Feature Flag Rollouts and Kill Switches
Design staged flag rollouts, verify rollback paths, and audit flag debt with explicit defaults, gates, and retirement plans.
Microcontroller Firmware and RTOS Engineering
Build and debug MCU firmware with bounded ISR/RTOS timing, safe DMA ownership, and measured power budgets.