executing-plans
Use when you have a written implementation plan to execute in a separate session with review checkpoints
When & Why to Use This Skill
The Executing Plans skill is designed to bridge the gap between planning and implementation by providing a structured, batch-based execution framework. It ensures high-quality output through a rigorous process of plan review, incremental task execution, and mandatory verification checkpoints. By integrating human-in-the-loop feedback and automated testing (such as .NET build and test cycles), it minimizes errors in complex workflows and maintains architectural integrity throughout the development lifecycle.
Use Cases
- Technical Implementation: Executing complex software feature plans by breaking them into manageable batches with automated build and test verifications.
- Incremental Refactoring: Managing large-scale codebase migrations or refactoring tasks that require frequent checkpoints to ensure system stability.
- Structured Task Management: Handling multi-session development projects where progress must be documented, verified, and reviewed by an architect at specific milestones.
- Quality Assurance Integration: Running systematic test plans where each phase of execution requires reporting results and obtaining human approval before proceeding.
| name | executing-plans |
|---|---|
| description | Use when you have a written implementation plan to execute in a separate session with review checkpoints |
Executing Plans
Overview
Load plan, review critically, execute tasks in batches, report for review between batches.
Core principle: Batch execution with checkpoints for architect review.
Announce at start: "I'm using the executing-plans skill to implement this plan."
The Process
Step 1: Load and Review Plan
- Read plan file
- Review critically - identify any questions or concerns about the plan
- If concerns: Raise them with your human partner before starting
- If no concerns: Create TodoWrite and proceed
Step 2: Execute Batch
Default: First 3 tasks
For each task:
- Mark as in_progress
- Follow each step exactly (plan has bite-sized steps)
- Run verifications as specified
- Mark as completed
For C# .NET Projects (ECTSystem):
# After each task in plan, run verification
# Build the solution to catch compilation errors early
dotnet build ECTSystem.sln /property:GenerateFullPaths=true /consoleloggerparameters:NoSummary /m:8
# Run tests to verify functionality
dotnet test AF.ECT.Tests --filter "YourTestNamePattern" --no-build
# Check for warnings that become errors
dotnet build ECTSystem.sln /property:GenerateFullPaths=true /p:TreatWarningsAsErrors=true
Step 3: Report
When batch complete:
- Show what was implemented
- Show verification output
- Say: "Ready for feedback."
For C# .NET Projects (ECTSystem):
Example report:
Completed Tasks 1-3:
Task 1: Added GetUserCount gRPC method
- Files modified: AF.ECT.Shared/Protos/workflow.proto, AF.ECT.Server/Services/WorkflowServiceImpl.cs
- Tests: 2/2 passing (GetUserCount_WithValidRequest_ReturnsCount, GetUserCount_WithInvalidStatus_ThrowsRpcException)
- Build: ✓ 0 errors, 0 warnings
Task 2: Implemented User audit logging
- Files modified: AF.ECT.Server/Interceptors/AuditInterceptor.cs, AF.ECT.Data/EctDbContext.cs
- Tests: 4/4 passing (AuditInterceptor_LogsMethodCall, AuditInterceptor_IncludesCorrelationId, etc.)
- Build: ✓ 0 errors, 0 warnings
Task 3: Added migrations for audit tables
- Files modified: AF.ECT.Database/dbo/Tables/AuditLogs.sql
- Tests: Migration tested via EF Core context (3 tables created successfully)
- Build: ✓ 0 errors, 0 warnings
Verification Results:
$ dotnet test AF.ECT.Tests --filter "Audit|UserCount" --no-build
9 tests passed, 0 failed
$ dotnet build ECTSystem.sln
Build complete with 0 errors, 0 warnings
Ready for feedback.
Step 4: Continue
Based on feedback:
- Apply changes if needed
- Execute next batch
- Repeat until complete
Step 5: Complete Development
After all tasks complete and verified:
- Announce: "I'm using the finishing-a-development-branch skill to complete this work."
- REQUIRED SUB-SKILL: Use superpowers:finishing-a-development-branch
- Follow that skill to verify tests, present options, execute choice
When to Stop and Ask for Help
STOP executing immediately when:
- Hit a blocker mid-batch (missing dependency, test fails, instruction unclear)
- Plan has critical gaps preventing starting
- You don't understand an instruction
- Verification fails repeatedly
Ask for clarification rather than guessing.
When to Revisit Earlier Steps
Return to Review (Step 1) when:
- Partner updates the plan based on your feedback
- Fundamental approach needs rethinking
Don't force through blockers - stop and ask.
Remember
- Review plan critically first
- Follow plan steps exactly
- Don't skip verifications
- Reference skills when plan says to
- Between batches: just report and wait
- Stop when blocked, don't guess