Spec Driven Implementation

Spec-Driven Implementation Skill

Purpose

Execute the planning artifacts produced by spec-driven-planning. The skill auto-detects which mode the feature is in and runs the matching playbook:

• Quick mode (plan.md exists) — follow bite-sized steps, run verification commands, commit. No RED-GREEN-REFACTOR. Tests welcome but not mandated.
• Full mode (tasks.md exists) — strict TDD per task: RED → GREEN → REFACTOR, traceability to REQ-### IDs, quality gates between tasks.

Activation Triggers

• User says "implement this feature", "let's code this", "execute the plan"
• User mentions "tasks", "TDD", or "execution"
• User uses /dev-workflow:spec with tasks or execute
• Planning phase is complete

Mode Detection (DO THIS FIRST)

Before doing anything, find the active feature folder and detect mode:

1. ls docx/features/  →  pick most recent or user-named feature
2. Inside docx/features/[NN-name]/, look for:
   - plan.md only                          → QUICK MODE
   - tasks.md only                         → FULL MODE
   - tasks.md + plan.archived.md           → FULL MODE (post-upgrade; archive is informational, ignored for routing)
   - plan.md + tasks.md (no archive)       → STOP AND ASK: ambiguous state.
                                              Likely a hand-conversion in progress.
                                              Ask user which file is authoritative.
                                              Do NOT auto-pick — risk of clobbering.
   - neither                               → ERROR: "No plan/spec found. Run /dev-workflow:spec first."

plan.archived.md is informational only and never affects mode detection — a clean post-upgrade state has tasks.md + plan.archived.md and routes to Full.

Announce the mode:

"Detected [Quick/Full] mode from docx/features/[NN-name]/[plan.md|tasks.md]. Running [Quick/Full] execution."

Then run the matching section below.

Quick Mode Execution

Source: docx/features/[NN-name]/plan.md

Philosophy: Behavior-based verification, not test-suite-based. The plan already tells you exactly what to do.

Pre-flight

1. Read plan.md end-to-end
2. Review critically — raise concerns BEFORE starting:
   • Are file paths real? (sample-check 1-2 with Read)
   • Do the steps actually solve the Goal?
   • Are there obvious gaps?
3. If concerns: surface them to the user, wait for guidance
4. If clean: proceed

Per-task loop

For each task in the plan:

1. Mark task as in-progress — Edit the Progress section: - [ ] Task N → - [→] Task N
2. For each step in the task (steps are themselves checkboxes - [ ] in the template):
   • Execute the action (write code from the plan, run the command, etc.)
   • Run the verification command shown in the plan
   • Compare actual output to "Expected:" — if mismatch, stop and ask, don't guess
   • Edit plan.md to flip the step's - [ ] → - [x] immediately after completing it
3. Commit using the message from the plan (the commit step itself is one of the checkboxed steps)
4. Mark task complete — Edit the Progress section: - [→] Task N → - [x] Task N

Quick mode rules

• No RED-GREEN-REFACTOR enforcement. If the plan lists a test step, do it. If not, don't invent one.
• Opting into TDD per-task is allowed. If a specific task warrants a test, the plan author adds an explicit test step (e.g., "1. Write failing test in tests/foo.test.ts ... 2. Verify: npm test foo → expect FAIL"). Execute it as ordinary verification. No test step in the plan = no test required.
• Don't expand scope. The plan is the contract. New ideas → tell the user, don't silently add tasks.
• Use Edit on plan.md, don't just announce progress. Mark step-level checkboxes (- [ ] → - [x]) as you complete each numbered step within a task — this preserves resumability if interrupted mid-task.
• Stop when blocked. Failing verification, missing file, ambiguous step → ask.
• No quality gate between tasks beyond what the plan specifies. (Optionally invoke dev-workflow:review before commits if the user requested it.)

Quick mode completion

When all tasks are [x]:

1. Update Status: line in plan.md to Complete
2. Run any final verification listed in the plan
3. Report:

Quick implementation complete: [Feature Name]

- Tasks: N/N
- Commits: M
- Files changed: K

Plan: docx/features/[NN-name]/plan.md
Ready for PR or further work?

Full Mode Execution

Source: docx/features/[NN-name]/{requirements.md, design.md, tasks.md}

Prerequisites Check

• requirements.md complete (EARS, REQ-### IDs)
• design.md complete and approved
• tasks.md exists (Phase 4 done) OR run Phase 4 below

If tasks.md is empty/missing, run Phase 4 first.

Pre-flight Review (before Phase 4)

Before breaking design into tasks or executing existing tasks, critically review the planning artifacts:

1. Are referenced file paths real? Sample-check 1-2 paths from design.md with Read. Hallucinated paths are a strong signal the design wasn't grounded in the actual codebase.
2. Do the requirements actually match what the user asked for? Re-read the original request and check requirements.md for drift.
3. Does the design solve the requirements? Each REQ-### should map to at least one design component.
4. Are there obvious gaps? Missing error handling, undefined interfaces, hand-waved sections.

If concerns: STOP. Surface them to the user BEFORE writing tasks or code. Don't push through hoping it'll work out.

🗣 Say: "Pre-flight: [N concerns found / clean — proceeding to Phase 4]."

Phase 4: Task Breakdown (TDD Focus)

Goal: Break design into small, testable tasks following Red-Green-Refactor

Task Structure:

Each task follows TDD cycle:

[ ] Task N: [Description]
    [ ] RED: Write failing test for [functionality]
    [ ] GREEN: Implement minimal code to pass test
    [ ] REFACTOR: Clean up and optimize

    Acceptance Criteria:
    [ ] [Specific criterion 1]
    [ ] [Specific criterion 2]
    Linked Requirements: REQ-###, REQ-###

Traceability Rules:

• Every task must list the requirement IDs it satisfies (from requirements.md)
• All requirements must appear in at least one task
• Repeat IDs across tasks if a requirement spans multiple tasks

Task Sizing: 30-60 minutes each. If longer, break into subtasks. Each task must be independently testable and produce working, tested code.

UltraThink Before Task Breakdown: Before breaking design into tasks, activate deep thinking if design involves complex algorithms, unclear integration points, multiple strategies, or non-trivial edge cases.

"Let me ultrathink the implementation strategy before breaking this into tasks."

Questions to ultrathink:

• What's the simplest implementation that satisfies requirements?
• Where are the hidden complexities?
• What assumptions might break during implementation?
• How will we test each component in isolation?

Task Categories:

1. Component Tasks - Individual components implementation
2. Integration Tasks - Connect components, test interactions, verify data flow
3. Error Handling Tasks - Error scenarios, edge cases, error messages
4. Documentation Tasks - Docstrings, README updates, API docs
5. Final Verification Tasks - Code review, performance, security, manual testing

Output: Update docx/features/[NN-feature-name]/tasks.md with implementation approach summary, organized task list with checkboxes, acceptance criteria, and notes.

"Tasks defined with TDD cycle. Ready to begin implementation?"

Phase 5: Execution (Full Mode TDD)

Goal: Execute tasks systematically with quality gates

For each task:

1. Mark Task as In Progress - Edit tasks.md: [ ] Task N → [→] Task N, mark RED as [→]

2. RED Phase - Write failing test, verify failure, Edit to check off RED [x], commit: test: Add test for [functionality]

3. GREEN Phase - Write minimal implementation, all tests must pass, Edit to check off GREEN [x], commit: feat: Implement [functionality]

4. REFACTOR Phase - Clean up code, tests still passing, Edit to check off REFACTOR [x], commit: refactor: Optimize [component]

5. Mark Task Complete - Edit: [→] Task N → [x] Task N, verify acceptance criteria checked, update Progress Summary

Task Tracking Protocol

CRITICAL: Use Edit tool to update tasks.md - don't just announce progress.

Start Phase 5
    ↓
Edit: Status "Not Started" → "In Progress"
    ↓
For each task:
    Edit: [ ] Task N → [→] Task N
    Edit: [ ] RED → [→] RED
    Write failing test
    Edit: [→] RED → [x] RED, [ ] GREEN → [→] GREEN
    Implement code
    Edit: [→] GREEN → [x] GREEN, [ ] REFACTOR → [→] REFACTOR
    Refactor code
    Edit: [→] REFACTOR → [x] REFACTOR, [→] Task N → [x] Task N
    Edit: Update Progress Summary
    ↓
Next task or finish
    ↓
Edit: Status "In Progress" → "Complete"

Announcing progress is NOT updating files. Always use Edit tool to modify tasks.md, then announce.

Progress Summary Maintenance

Keep the Progress Summary section in tasks.md synchronized after every task:

- Total Tasks: 10
- Completed: X/10
- In Progress: Task N - [description]

Status Checkpoints

Every 2-3 completed tasks:

Checkpoint Update:
- Tests: [N/N] passing
- Type check: No errors
- Lint: Clean
- Completed tasks: [X/Y]
- Next: [Next task description]
[Confidence: X.X]

Auto-Trigger Code Quality Review

Before each commit, invoke dev-workflow:review to review changes and address critical findings.

Integration Skills:

• dev-workflow:test-driven-development - Strict TDD guidance
• dev-workflow:git-workflow - Smart commits
• dev-workflow:review - Pre-commit reviews

Completion Criteria:

• All tasks checked off
• All tests passing
• No linter errors
• No type errors
• Code review passed
• Documentation updated
• Manual testing complete

Final Output:

Feature Complete: [Feature Name]

Summary:
- Total tasks: [N]
- Tests added: [N]
- Test coverage: [X%]
- Commits: [N]
- Files changed: [N]

Ready to create PR or merge to main?

TDD Principles

Red-Green-Refactor Cycle

• RED: Write a test that describes what the code should do. It MUST fail because the code doesn't exist yet.
• GREEN: Write MINIMUM code to make the test pass. Don't optimize, just make it work.
• REFACTOR: Improve the code (remove duplication, improve naming, simplify). Run tests after each change.

TDD Rules

DO: Write test before code (always RED first), write smallest test possible, run tests frequently, commit at each phase, keep tests fast and independent.

DON'T: Write production code without a failing test, write multiple tests before implementation, skip refactor phase, test implementation details (test behavior), let tests become slow or flaky.

Example: Task Breakdown

User: "/dev-workflow:spec tasks"

Assistant:

Activating Phase 4: Task Breakdown

Found feature: docx/features/01-user-authentication/
Reading design to create TDD tasks...

[ ] Task 1: User model with password hashing
    [ ] RED: Test password hashing on save
    [ ] GREEN: Implement bcrypt hashing
    [ ] REFACTOR: Extract hashing to utility
    Acceptance Criteria:
    [ ] Passwords are hashed before storage
    [ ] Original password is not stored
    Linked Requirements: REQ-001, REQ-002

[...more tasks...]

Tasks defined. Ready to begin implementation?

Back to Planning

If during implementation you discover missing requirements, design changes needed, or unclear specifications:

"Implementation blocked: [issue]. Need to revisit [requirements/design]. Use /dev-workflow:spec [requirements/design] to update planning documents."

Invoke dev-workflow:spec-driven-planning to update planning documents before continuing.

Critical Reminders

• NEVER write production code without a failing test first
• ALWAYS use Edit tool to update tasks.md checkboxes - don't just announce progress
• ALWAYS update Progress Summary after each task completion
• Update Status header when starting/completing implementation
• Follow TDD cycle religiously (RED → GREEN → REFACTOR)
• Provide checkpoint updates every 2-3 tasks
• Stop and return to planning if design issues discovered