Speckit Implement

Spec-Kit Implement

Execute the implementation by processing tasks from tasks.md in phases. Fourth step in spec-kit workflow.

When to Use

• After creating tasks.md with speckit-tasks
• Ready to start implementation
• Want to execute tasks systematically

Execution Workflow

1. Setup: Run .specify/scripts/bash/check-prerequisites.sh --json --require-tasks --include-tasks to get FEATURE_DIR and AVAILABLE_DOCS
2. Check checklists status (if FEATURE_DIR/checklists/ exists):
   • VALIDATION GATE: Scan all checklist files and count total, completed, incomplete items
   • Create status table showing pass/fail for each checklist
   • If any checklist incomplete: STOP and ask user if they want to proceed anyway
     • User must explicitly approve to continue
   • If all complete: Automatically proceed to implementation
3. Load implementation context:
   • REQUIRED: tasks.md (complete task list), plan.md (tech stack, architecture)
   • IF EXISTS: data-model.md, contracts/, research.md, quickstart.md
4. Project setup verification:
   • Create/verify ignore files based on actual project setup (gitignore, dockerignore, eslintignore, etc.)
   • Detect technology from plan.md and add appropriate patterns
   • Only append missing critical patterns to existing ignore files
5. Parse tasks.md structure:
   • Extract task phases, dependencies, details (ID, description, file paths)
   • Identify parallel markers [P] and sequential vs parallel execution rules
6. Execute implementation following task plan:
   • Phase-by-phase execution (complete each phase before next)
   • Respect dependencies (sequential tasks in order, parallel tasks [P] can run together)
   • Follow TDD approach if tests exist (execute test tasks before implementation)
   • File-based coordination (tasks on same files run sequentially)
   • Validation checkpoints after each phase
7. Implementation execution order:
   • Setup first (initialize structure, dependencies, configuration)
   • Tests before code (if test tasks exist)
   • Core development (models, services, CLI commands, endpoints)
   • Integration work (database, middleware, logging, external services)
   • Polish and validation (unit tests, optimization, documentation)
8. Progress tracking and task completion:
   • Report progress after each completed task
   • Halt execution if non-parallel task fails
   • For parallel tasks, continue with successful ones, report failures
   • CRITICAL: Mark completed tasks immediately with [X] in tasks.md
     • Change - [ ] T001 to - [X] T001 after completion
     • Do NOT batch task updates - mark each one as you complete it

Key Points

• Load only required context - Don't read every artifact, only what's needed
• Respect task dependencies - Sequential vs parallel execution
• Mark tasks complete immediately with [X] in tasks.md
• Create ignore files based on detected technology (from plan.md)
• Phase-by-phase execution - Complete each phase before moving to next
• Error handling - Halt on critical failures, report parallel task failures
• Checklist validation - Warn if checklists incomplete before starting

Next Steps

After implementation:

• Manual testing and QA
• Code review
• Deployment planning

See Also

• speckit-tasks - Break plan into actionable tasks
• speckit-checklist - Generate custom validation checklists
• speckit-analyze - Validate cross-artifact consistency